Induction of cell cycle arrest and apoptosis by a combined treatment with 13-cis-retinoic acid, interferon-α2a, and α-tocopherol in squamous cell carcinoma of the head and neck

ORIGINAL ARTICLE

INDUCTION OF CELL CYCLE ARREST AND APOPTOSISBY A COMBINED TREATMENT WITH 13-CIS-RETINOIC ACID,INTERFERON-a2A, AND a-TOCOPHEROL IN SQUAMOUSCELL CARCINOMA OF THE HEAD AND NECK

Xin Zhang, MD, PhD, Zhuo (Georgia) Chen, PhD, Fadlo R. Khuri, MD, Dong M. Shin, MD

Winship Cancer Institute, Emory University School of Medicine, Atlanta, Georgia 30322.E-mail: [email protected]

Accepted 12 July 2006Published online 12 December 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/hed.20525

Abstract: Background. We have previously conducted

phase II trials with a combination of 13-cis-retinoic acid (13-

cRA), interferon-a2a (IFN-a2a), and a-tocopherol (a-TF) in

patients with advanced oral premalignant lesions and locally

advanced head and neck cancer in the adjuvant settings and

achieved promising outcomes. The present study was con-

ducted in vitro to elucidate the mechanisms of anti-tumor activity

of this 3-drug combination in squamous cell carcinoma of the

head and neck (SCCHN).

Methods. Five SCCHN cell lines were treated with 13-cRA,

IFN-a2a, and a-TF as single agents or 2- to 3-drug combinations

for 72 hours. Inhibition of cell growth and cell cycle progression

and induction of apoptosis by the treatments were evaluated.

Results. Our results demonstrated that although each sin-

gle-agent and 2-drug combination showed a certain level of cell

growth inhibition, the 3-drug combination apparently further

inhibited cell growth in comparison to any single agents and

2-drug combinations in the 5 SCCHN cell lines. Cell cycle analy-

sis on Tu212 and 886LN cells by flow cytometry exhibited signifi-

cant accumulation of the cells at S phase in the 3-drug combina-

tion. On the other hand, Annexin-V binding assay demonstrated

that the 3-drug combination induced more profound apoptosis

than any of the single agents or 2-drug combinations. In parallel,

proteolytic cleavages of pro-caspase-8, -9, -3 and poly (ADP

ribose) polymerase as well as caspase-3 activity induced by the

3-drug treatment were observed.

Conclusions. Our data suggests that 3-drug combination

biochemopreventive regimen has cooperative inhibitory effect

on the growth of SCCHN cells. Both cell cycle arrest and apopto-

sis contribute to cell growth inhibition of this 3-drug combina-

tion therapy. VVC 2006 Wiley Periodicals, Inc. Head Neck 29:

351–361, 2007

Keywords: squamous cell carcinoma of the head and neck;

retinoic acid; interferon-a2a; a-tocopherol

The rising incidence, morbidity, and mortality ofsquamous cell carcinoma of the head and neck(SCCHN) have presented major challenges to thehead and neck oncology community. Despite theimprovement in conventional therapies, includingsurgery, radiotherapy, and chemotherapy, theoverall survival rate of the disease has remainedlargely unchanged over the decades. In the pastseveral years, however, novel approaches to che-moprevention and adjuvant therapy have beenactively tested preclinically and clinically inSCCHN. Numerous studies have revealed thatretinoic acids (RAs) and interferon-a (IFN-a) have

Correspondence to: D. M. Shin

VVC 2006 Wiley Periodicals, Inc.

Contract grant sponsor: NCI; contract grant numbers: CA75603,CA101244.

Cell Cycle Arrest and Apoptosis by Biochemoprevention HEAD & NECK—DOI 10.1002/hed April 2007 351

significant anti-tumor activities as single-agentsboth in vitro and in vivo,1–5 and that these 2 drugsact synergistically when used in combination ina variety of cancer types, including SCCHN.6–10

Another chemopreventive agent, a-tocopherol (a-TF), is believed to reduce toxicities and enhancethe activity of 13-cis-retinoic acid (13-cRA).11 Ourbiochemopreventive study in patients with ad-vanced premalignant lesions (moderate or severedysplasia) of the upper aerodigestive tract showedthat the 3-drug combination of 13-cRA, IFN-a, anda-TF achieved a substantial rate of pathologicresponses. The reversed pathologic responses werecorrelated with p53 status and expression of chro-mosome polysomy.12,13 A phase II trial conductedby our group also evaluated the clinical efficacy ofthis 3-drug combination as bio-adjuvant therapyagainst locally advanced (stage III/IV) SCCHNand showed median 1- and 2-year overall survivalrates of 98% and 91%, respectively, and disease-free survival rates of 91% and 84%, respectively.This combination regimen also achieved low inci-dence of second primary tumors and tolerable tox-icities.14 Long-term follow up at a median of 49.4months revealed that overall survival rates at 1,3, and 5 years were 98%, 89%, and 81%, respec-tively. Disease-free survival rates at 1, 3, and 5years were 89%, 82%, and 80%, respectively, sug-gesting that the effect of this bioadjuvant therapyis also long lasting.15

As potent chemopreventive agents, 13-cRA andIFN-a have been extensively studied individuallyfor their activities in inhibiting cell proliferationand inducing cell differentiation and apoptosis.Upon entry into cells, RAs bind to their nuclear re-ceptors (RARs or RXRs), which then activate theexpression of target genes by binding to specificRA-responsive elements and mediate a variety ofbiological functions.16–18 The biological activitiesof IFN-a are achieved by specific receptor binding,which subsequently activates signal transducerand transcription activator (STAT)-1 and -2 signaltransduction pathways and induces transcriptionof genes that contain the IFN-stimulated responseelement in their promoters.19–21 Though RAs andIFNs have been widely shown to demonstrate bothsingle-agent activities and combined synergisticeffects on the modulation of cell proliferation, dif-ferentiation, apoptosis, and angiogenesis,6,8,10,20

the biological activities of the 3 drugs of 13-cRA,IFN-a, and a-TF in combination have not beenwell characterized.

The promising results in clinical chemopreven-tion and adjuvant therapy have prompted us to

further investigate the mechanisms of action ofthe 3-drug combination at the cellular and molec-ular levels. The present in vitro study evaluatedthe cell growth effects of 13-cRA, IFN-a2a, anda-TF and their combinations on SCCHN cells.Our results demonstrated that the 3-drug combi-nation inhibited the growth of all 5 SCCHN celllines more potently than any of the single agentsor 2-drug combinations. We also found that suchenhanced inhibitory effects were attributed toboth cell cycle arresting and increased apoptosis.

MATERIALS AND METHODS

Cell Lines. Five SCCHN cell lines (Tu177, Tu212,886LN, SQCCY1, and 38) were used for this study.The Tu177 cell line was established from the lar-ynx, and Tu212 was established from a primaryhypopharyngeal tumor. These 2 cell lines wereprovided by Dr. Gary L. Clayman (The Universityof Texas M. D. Anderson Cancer Center, Houston,TX). The 886LN cell line was established from alymph node metastasis of squamous cell carci-noma of the larynx. SQCCY1 cell line was fromSCC in oral cavity, and cell line 38 was establishedfrom tonsil fossa. These 3 cell lines were obtainedfrom Dr. Shi-Yong Sun (Winship Cancer Institute,Emory University, Atlanta, GA). They were main-tained in Dulbecco’s modified Eagle’s medium(DMEM)/F12 (1:1) medium supplemented with5% heat-inactivated fetal bovine serum and anti-biotics (streptomycin, penicillin G, and amphoter-icin B) in a 378C, 5% (carbon dioxide) CO2 humidi-fied incubator.

Reagents. Agents of 13-cRA (Sigma Chemical,St. Louis, MO) and a-TF (Sigma Chemical) werefirst dissolved in dimethyl sulfoxide (DMSO) asstock solutions, and then diluted in DMEM/F12medium immediately before use. The final concen-tration of DMSOwas 0.1% or less. IFN-a2a (Accu-rate Chemical & Scientific, New York, NY) wasdirectly dissolved in DMEM/F12 medium immedi-ately prior to use.

Cell Growth Inhibition Assay. To test the effects of13-cRA, a-TF, and IFN-a2a on the growth ofSCCHN cells as single agents or in combination,cells were seeded in 96-well plates at a density of4000 cells/well overnight prior to drug treatment.Subsequently, drugs were added as single agentsat various concentrations (0–200 lM for 13-cRA,0–50,000 U/mL for IFN-a2a, and 0–100 lM for

352 Cell Cycle Arrest and Apoptosis by Biochemoprevention HEAD & NECK—DOI 10.1002/hed April 2007

a-TF), followed by incubation at 378C and 5% CO2

for 72 hours. In the following experiments, thesingle concentrations for each drug were carefullychosen to achieve 10% to 30% of inhibition in eachcell line (1 lM for 13-cRA, 200 U/mL for IFN-a2a,15 lM for a-TF). Afterward, each cell line wastreated with single, double, or triple drug combi-nations using the chosen fixed concentrations for72 hours. Cell growth inhibition was measured bydetermining cell density with sulforhodamine B(SRB) assay.22 Percentage of inhibition was deter-mined by comparison of cell density in the drug-treated cells with that of the control cells. Allexperiments were repeated at least 3 times.

Flow Cytometry Analysis. The effects of each ofthe single agents and combinations on cell cyclearrest and apoptosis were analyzed in 2 cell linesTu212 and 886LN. Cells were trypsinized andwashed in 13 phosphate buffered saline (PBS)before flow cytometry analysis was performed.For cell cycle analysis, cells were fixed in 70%ethanol at 48C for 2 hours, followed by awashwith13 PBS. Cells were then stained with (propidiumiodide) PI/RNase staining buffer (BD PharMin-gen, San Diego, CA) for 15 minutes at room tem-perature. For apoptosis assay, drug-treated cellswere trypsinized and washed with PBS and resus-pended in Annexin-binding buffer (BD PharMin-gen). Cells were then stained with both AnnexinV-phycoerythrin (Annexin V-PE; BD PharMin-gen) and 7-amino-actinomycin (7-AAD; BD Phar-Mingen) for 15 minutes at room temperature. Thestained samples for both the cell cycle and apopto-sis assay were measured using a fluorescence-activated cell sorting caliber bench-top flow cy-tometer (Becton Dickinson, Franklin Lakes, NJ).The data were analyzed using FlowJo software(Tree Star, Inc., Ashland, OR). The experimentswere repeated at least 3 times independently.

Western Blot Analysis. Whole cell lysates wereextracted from drug-treated cells using a lysisbuffer. Fifty micrograms of protein was separatedon 8% to 15% sodium dodecyl sulfate polyacryl-amide gel electrophoresis (SDS-PAGE) gel, trans-ferred onto a polyvinylidene fluoride membrane(Millipore Co., Bedford MA), and immunostainedwith specific antibodies, including anti-mouse cas-pase-3 (Imgenex, San Diego, CA), anti-mouse cas-pase-8, anti-rabbit caspase-9, and anti-rabbit pol-y(ADP)-ribose polymerase (PARP) (Cell Signal-ing, Beverly, MA). Immunostaining with anti-b-

actin antibody (Sigma Chemical) was used assample loading control. Immunostained proteinbands were detected with an enhanced chemilu-minescense kit (ECL, Ambersham, Buckingham-shire, UK). The experiments were repeated atleast 3 times.

Caspase-3 Activity Assay. This experiment wasperformed in 96-well plate. Ten micrograms ofprotein from cell lysates and 1 lL of 5 mM CPP32(a 32-kDa putative cysteine protease) substrate(DEVD-AFC, BIOMOL Research LaboratoriesInc., PA) were mixed with caspase reaction buffercontaining 20 mM Pipes, 100 mM NaCl, 10 mmDTT, 1 mM EDTA, 0.1% CHAPS, and 10% Su-crose in each well. The final volume of reactionsystemwas 100 lL. The reaction systemwas incu-bated at 378C for 1 hour. The results were mea-sured by a fluorescence microplate reader (SPEC-TRAmax GEMINI XS Dual-scanning MicroplateSpectrofluorometer, Sunnyvale, CA). A negativecontrol was performed to evaluate the specificityof the assay by pretreating the drug-treated sam-ples with CPP32 inhibitor (DEVD-CHO, BIOMOLResearch Laboratories Inc., PA) at 378C for 30min-utes prior to the incubation with the fluorescentsubstrate. The experiments were repeated 2 times.

Statistical Analysis. Cell growth inhibition assay(for 3-drug combination) was evaluated using 2-sided and 2-sample equal variance Student’s ttests. Cell cycle, apoptosis (annexin-V binding),and caspase-3 activity assays were analyzed with2-sided paired Student’s t tests. In all analyses,p < .05 was considered to be statistically signi-ficant.

RESULTS

Effects of 13-cRA, IFN-a2a, and a-TF as Single

Agents and Their Combinations on the Growth of

SCCHN Cell Lines. To study the sensitivity ofSCCHN to 13-cRA, IFN-a2a, and a-TF, we treated5 SCCHN cell lines, Tu212, Tu177, 886LN,SQCCY1, and 38, with 13-cRA (0–200 lM), IFN-a2a (0–50,000 U/mL), and a-TF (0–100 lM) for 72hours. Cell growth inhibition assay showed thatall 3 compounds of 13-cRA, IFN-a2a, and a-TF assingle agents inhibited the growth of the fiveSCCHN cell lines in a dose-dependent manner(Figures 1A–1C). However, only less than 40% ofcell growth inhibition was achieved, even though


a very high concentration of IFN-a2a (50,000 U/mL) was administered to the cells, indicating thatthe effect of growth inhibition of IFN-a2a as a sin-gle agent is not much effective for SCCHN.

To test whether a combined treatment of 13-cRA, IFN-a2a, and a-TF would achieve a highergrowth inhibitory effect than each single agent orany 2-drug combination, the fixed concentrationsof each single agent were chosen for the combinedtreatment of SCCHN cell lines. The concentra-tions for the 3 drugs were as follows: 1 lM 13-cRA,200 U/mL IFN-a2a, and 15 lM a-TF, which arethe clinically relevant concentrations. Cell growthinhibition assay showed that each single agent atthe fixed concentration slightly inhibited growthof the 5 SCCHN cell lines. More effective inhibi-tion of cell growth was observed in each 2-drugcombination of 13-cRA and IFN-a2a, IFN-a2a anda-TF, or 13-cRA and a-TF. Furthermore, the 3-drug combination showed even more potent inhi-bition of the 5 SCCHN cell lines than any singleagents or 2-drug combinations (p < .05) (see Fig-ure 2).

Cell Cycle Arrest Induced by the Three-Drug Combi-

nation. To understand the mechanisms of cellgrowth inhibition observed by the 3-drug combi-nation, we further investigated the effects of the3-drug combination on cell cycle progression using2 representative SCCHN cell lines, Tu212 and886LN. As shown in Figure 3 and Table 1, cellcycle analysis of Tu212 cells at 48 hours showedan increased accumulation of S phase in the 3-drugcombination group (42.1%) compared to the controlgroup (28.7%). The 3-drug combined treatmentfor 72 hours slightly further induced S phasearrest (45.5%). In parallel, a remarkable decreaseof cell populations at the G2/M phase was ob-served in the 3-drug combination group as com-pared to the control group at both 48 and 72 hours.Similar results were also observed in the 886LNcell line. Differences in cell population at S phasebetween the 3-drug combined treatment and thecontrol group in both Tu212 and 886LN cells werestatistically significant (p < .05). These findingsindicate that cell cycle arrest at S phase maybe one of the mechanisms for the observed cellgrowth inhibition by the 3-drug combinationtreatment.

Apoptotic Effects of the Combined Treatment of 13-

cRA, IFN-a2a, and a-TF on SCCHN. To examinewhether growth inhibition by 13-cRA, IFN-a2a,

FIGURE 1. Effects of 13-cis-retinoic acid (13-cRA), interferon-a(IFN-a), and a-tocopherol (a-TF) on growth of squamous cell

carcinoma of the head and neck (SCCHN) cell lines. SCCHN cell

lines Tu212, Tu177, 886LN, SQCCY1, and 38 were treated with

13-cRA (0–200 lM) (A), IFN-a 2a (0–50,000 U/mL) (B), and a-TF(0–100 lM) (C) as single agents for 72 hours as described in

Materials and Methods. The results showed that 13-cRA, IFN-a,and a-TF inhibited growth of SCCHN cells in a dose-dependent

manner.


a-TF, and their combinations was also attribut-able to programmed cell death in SCCHN, we fur-ther investigated the effects of these treatmentson the apoptosis of Tu212 and 886LN cells. BothTu212 and 886LN cells were treated with 13-cRA(1 lM), IFN-a2a (200 U/mL), and a-TF (15 lM) aseither single agents or with their combinations for72 hours. In the 886LN cells, flow cytometry anal-ysis using annexin V staining demonstrated that13-cRA, IFN-a2a, and a-TF as single agentsslightly induced apoptosis (15.8% with 13-cRA,16%with IFN-a2a, and 14.3%with a-TF) as com-pared with the control group (8.2%); the 2-drugcombinations resulted in further apoptosis (22.5%with R-I, 27.5% with R-T, and 18.4% with I-T).Most importantly, the 3-drug combination morepotently induced apoptosis (35.8%) than any ofthe single agents or 2-drug combinations (Figures4A and 4B). A representative apoptosis result waspresented in Figure 4A. A similar result wasobserved in Tu212 cells (Figure 4B). There wassignificant difference between the 3-drug combina-tion and 2-drug combination of IFN-a2a and a-TFin 886LN cells (p ¼ .05). Significant differenceswere also found in the 3-drug combination as com-pared with the 2-drug combination of 13-cRA þIFN-a2a (p ¼ .002) and 13-cRA þ a-TF in Tu212cells (p¼ .01).

Effects of the Combined Treatment of 13-cRA, IFN-

a2a, and a-TF on Activation of Apoptosis-Related

Proteins. To investigate the possible pathwaysinvolved in the 3-drug combinations induced apo-ptosis, we checked the status of activation of apo-ptosis-related proteins. Tu212 and 886LN cellswere treated with 13-cRA, IFN-a2a, and a-TF aseither single agents or in combinations for 72hours. Immunoblotting was employed to detectthe activation of caspase-8, -9, and -3, and PARP,which are the major cascade of proteins involvedin apoptosis-related pathways. Our results showedthat 2-drug combinations induced activation ofcaspase-8, -9, and -3, and PARP at different levelsas compared with the control group, althoughthe control group also showed basal levels of ac-tivated caspase-9 and PARP in Tu212 and 886LN,respectively. The 3-drug combination further in-creased the expression of activated caspase-9, -3,and PARP as compared with either single agentsor 2-drug combinations in the 886LN cells (seeFigure 5).

Because of the central role of caspase-3 inmediating cell apoptosis, we specifically assessedthe effects of the 3-drug combinations treatmenton caspase-3 activity. Tu212 and 886LN cells weretreated with either single agents or their combina-tions for 72 hours. Caspase-3 activity assay was

FIGURE 2. Effect of a combination of 13-cis-retinoic acid (13-cRA), interferon-a (IFN-a), and a-tocopherol (a-TF) on growth of SCCHN

cell lines. Cell lines Tu212, Tu177, 886LN, SQCCY1, and 38 were treated with single agents of 13-cRA (1 lM), IFN-a (200 IU/mL), a-TF(15 lM), double combination of 13-cRA and IFN-a (R-I), 13-cRA and a-TF (R-T), or IFN-a and a-TF (I-T), or triple combination of 13-cRA,

IFN-a, and a-TF (R-I-T) for 72 hours, as described in Materials and Methods. The result showed that treatment with the triple combination

further inhibited growth of SCCHN cells as compared with either single agents or 2-drug combinations. * Stands for the comparison

between the 3-drug combination and any single agent treatment or 2-drug combinations in each cell line (p< .05).


measured using a fluorescence microplate reader.The results demonstrated that both single agentsand 2-drug combinations induced certain levels ofcaspase-3 activity. However, the 3-drug combina-tion further increased the activity. There were sig-nificant differences between the 3-drug combina-tion and the 2-drug combinations in both Tu212and 886LN cells (except the comparison betweenthe 3-drug combination and the 2-drug combina-tion of 13-cRA þ a-TF, p ¼ .1) (see Figure 6). Theelevated caspase-3 activity was consistent withthe degrees of growth inhibition and apoptosis inboth Tu212 and 886LN cells, suggesting that cas-pase-3 may act as a mediator of apoptosis in

Tu212 and 886LN cells by this 3-drug combinationtreatment.

DISCUSSION

Retinoids at high dose have established efficacy inreversal of early premalignant lesions and in pre-venting second primary tumors in SCCHN. How-ever, mucocutaneous toxicity was significant andthe duration of responses was limited. A singleagent of 13-cRA at low dose had no effect on pre-venting secondary primary tumors in stage I/IIhead neck cancer.23,24 Therefore, a combination oflow-dose retinoids with other chemopreventive

FIGURE 3. Cell cycle arrest induced by a combination treatment of 13-cis-retinoic acid (13-cRA), interferon-a (IFN-a), and a-tocopherol (a-TF) in Tu212 and 886LN cell lines. Tu212 and 886LN cells were treated with a 3-drug combination of 13-cRA, IFN-a,and a-TF for 48 and 72 hours, and cell cycle arrest was measured using flow cytometry as described in Material and Methods. The

results demonstrated that the 3-drug combination apparently induced S phase arrest as compared with the nontreatment control.

(The numerical data and statistics were shown in Table 1).


agents is warranted to enhance its efficacy andreduce its toxicity. For this purpose, the resultspresented in this study demonstrated a coopera-tive inhibitory effect of combined treatments of13-cRA, IFN-a2a, and a-TF on the growth ofSCCHN cell lines. This inhibitory activity wasachieved through both cell cycle arrest and in-creased apoptosis.

Though anti-tumor activities of RAs and IFN-a have been extensively reported,2,3 the effects ofthe 3-drug combination of 13-cRA, IFN-a2a, anda-TF on cell growth have not been well studied.Our recent phase II clinical trial evaluated theclinical efficacy of a combination of 13-cRA, IFN-a2a, and a-TF as adjuvant therapy against locallyadvanced SCCHN. The results showed significantimprovement of both overall and disease-free sur-vival rates as compared with what has beenreported in the literature for similar patient co-horts,14,15 a promising observation that promptedus to conduct a mechanistic study on this 3-drugcombination. Previously, Sacks et al25–27 demon-strated slightly inhibitory effects of RAs and IFNsas single agents on the growth of SCCHN cells,886LN. a-TF has been used as a chemopreventiveagent against various cancer types.28 It is believedthat a-TF could reduce 13-cRA induced toxicityand enhance 13-cRA activity.11 On the other hand,a-TF itself is also able to inhibit tumor cell growthin a variant of cancer types.29–31 In this study, wecompared the growth inhibitory effects of the sin-gle agents of 13-cRA, IFN-a2a, and a-TF, andtheir combination as 2 drugs as well as 3 drugs on5 SCCHN cell lines (TU212, Tu177, 886LN,SQCCY1, and 38). Our results showed that all3 drugs as single agents inhibited the growth of all5 cell lines in a dose-dependent manner. Then wechose fixed concentrations of each single agent for

the combination study. Their concentrations wereas follows: 1 lM 13-cRA, 200 U/mL IFN-a2a, and15 lM a-TF. These selected concentrations arevery close to the peak plasma levels achievedwhen the therapeutic doses of these 3 drugs wereadministered to the patients.32–34 At the fixed con-centrations, each single agent achieved only mini-mal cell growth inhibition. Two-drug combinationsapparently inhibited cell growth better than thesingle-drug treatments. More importantly, the 3-drug combination significantly induced cell growthinhibition compared with any single agents or 2-drug combinations in all 5 SCCHN cell lines. Ourresults clearly demonstrated a cooperative effectof the 3-drug combination on cell growth inhibi-tion, suggesting that the addition of a-TF to thecombination of RAs and IFN-a is valuable.

To study the underlying mechanisms of the ob-served cell growth inhibition, we further analyzedthe effects of the 3-drug combination on cell cycleprogression. Previous studies showed that the3 drugs of 13-cRA, IFN-a2a, and a-TF as singleagents exerted different effects on cell cycledepending on the cancer type. A cell line SCC-15treated with 13-cRA exhibited cell cycle accumu-lation at S phase,35 while 13-cRA induced cellcycle arrest at G1 phase in leukemia cell lines.36

Similarly, IFN-a suppressed proliferation of chronicmyelogenous leukemia cells K562 by extendingcell cycle S phase,37 but it resulted in G1 phasearrest in malignant lymphoid cell lines.38 In pros-tate cancer, a-TF induced cell cycle arrest at G1phase in LNCaP cells, but cell cycle arrest at G2/M phase in PC3 cells.39 To avoid this complication,we simply compared the cell cycle effects betweenthe 3-drug combination and the control group inthis study of SCCHN cell lines. Our results re-vealed that the 3-drug combination obviouslyinduced cell cycle arrest at S phase compared withthe control in different time points in both Tu212and 886LN cell lines, indicating that the bioadju-vant therapy suppresses cell proliferation throughcell cycle arrest mainly at S phase.

A great deal of evidence demonstrated that allthree drugs of 13-cRA, IFN-a, and a-TF as singleagents can inhibit tumor growth through induc-tion of apoptosis in a variety of cancers.36,40,41 Toevaluate whether cell growth inhibition inducedby the 3-drug combination was also mediatedthrough apoptosis, both Tu212 and 886LN celllines were treated with a fixed concentration of 13-cRA (1 lM), IFN-a (200 U/mL), and a-TF (15 lM).Annexin-V binding assay revealed that the 3drugs as single agents slightly induced apoptosis.

Table 1. Cell cycle analysis.

Phase

TU212 886LN

Control R-I-T (p)* Control R-I-T (p)*

48 h

G0-G1 45.9 47.8 50.0 36.8

S 28.7 42.1 (0.04) 22.5 37.7 (0.01)

G2-M 24.7 10.1 24.2 24.8

72 h

G0-G1 47.5 40.6 60.7 47.6

S 32.2 45.5 (0.05) 20.4 38.5 (0.04)

G2-M 19.4 13.5 16.8 13.9

Abbreviations: R-I-T, 13-cRA þ IFN-a2a þ a-TF; 13-cRA, 13-cis-retinoicacid; IFN-a2a, interferon-a2a; a-TF, a-tocopherol.*p value was determined for comparison of S phase in control with thatin the 3-drug combination by t test from 3 data points.


However, the apoptotic cell population was some-what increased in the 2-drug combinations. Mostimportantly, the 3-drug combination showed

enhanced induction of apoptosis compared withany single agents or 2-drug combinations. Thesefindings suggest that the 3-drug combination is

FIGURE 4. Induction of apoptosis by a combination treatment of 13-cis-retinoic acid (13-cRA), interferon-a (IFN-a), and a-tocopherol(a-TF) in Tu212 and 886LN cell lines. Tu212 and 886LN cells were treated with single agents of 13-cRA (R, 1 lM), IFN-a (I, 200 IU/

mL), a-TF (T, 15 lM), 2-drug combinations of R-I, R-T, and I-T, or 3-drug combination of R-I-T for 72 hours. Apoptosis was detected

by flow cytometry as described in Materials and Methods. The 3-drug combination further induced apoptosis as compared with either

single agents or any 2-drug combinations. (A) Representative result for apoptosis induced by 3 drugs in combination. (B) Statistical

analysis results for 3 drugs-induced apoptosis in both Tu212 and 886LN cells. (*, p ¼ .002, .01, and .17 compared to 2-drug combina-

tions of R-I, R-T, and I-T, respectively, in Tu212 cells;. **, p ¼ .05, .07, and .07 compared to 2-drug combinations of R-I, R-T, and I-T,

respectively, in 886LN cells).


more effective to inhibit cell growth through in-duction of apoptosis than any of the single agentsor 2-drug combinations.

Small molecules usually induce apoptosisthrough 2 major pathways: extrinsic and intrin-sic pathways. The extrinsic pathway is initiatedwhen a death receptor ligand (ie, the Fas ligand)binds to its death receptor (ie, the Fas receptor).This interaction promotes the recruitment of

adapter molecules and results in the cleavage ofpro-caspase 8 to yield active caspase 8. Theintrinsic pathway is regulated by the perme-abilization of mitochondrial membranes. Mito-chondrial damage results in cytochrome c releaseand formation of the apoptosome, a multimericprotein complex containing caspase-9. Once boundto the apoptosome, caspase-9 is activated. Bothcaspase-8 and caspase-9 then activate down-

FIGURE 5. Inducing activation of the apoptotic proteins by 3 drugs in combination in Tu212 and 886LN cells. Tu212 and 886LN cells

were treated with 13-cRA (R, 1 lM), IFN-a (I, 200 IU/mL), a-TF (T, 15 lM,), 2-drug combinations of R-I, R-T, and I-T, or 3-drug combi-

nation of R-I-T for 72 hours. Fifty microgram protein was used for immunoblotting analysis using corresponding antibodies as

described in Materials and Methods. The 3-drug combination of R-I-T more effectively induced activation of caspase-8, -9, -3, and

PARP than either single agents or 2-drug combination.

FIGURE 6. Caspase-3 activity induced by 3 drugs in combination in Tu212 and 886LN cells. Tu212 and 886LN cells were treated with

13-cRA (R, 1 lM), IFN-a (I, 200 IU/mL), a-TF (T, 15 lM), 2-drug combinations of R-I, R-T, and I-T, or 3-drug combination of R-I-T for

72 hours. Caspase-3 activity was detected by a fluorescence microplate reader as described in Materials and Methods. The 3-drug

combination more potently induced caspase-3 activity than either single agents or 2-drug combinations (*, p ¼ .01, .01, and .005 com-

pared to 2-drug combinations of R-I, R-T, and I-T, respectively, in Tu212 cells; **, p ¼ .03, .1, and .005 compared to 2-drug combina-

tions of R-I, R-T, and I-T, respectively, in 886LN cells).


stream caspases, including caspase-3, in turn lead-ing to cleavages of their target proteins such asPARP, and eventually inducing apoptosis.42,43

Three drugs as single agents were reported toinduce apoptosis through both the extrinsic andintrinsic pathways. 13-cRA induces mitochondrialpermeability transition and release of cytochromec, suggesting a potential mitochondria-mediatedcell apoptosis.44 TRAIL and Fas are 2 of the majorcell death receptor ligands and receptors, respec-tively. Both IFN-a and a-TF can induce Fasexpression and enhance TRAIL-mediated apopto-sis by up-regulating caspase-8 activity in severalcancer types,45–48 indicating the extrinsic path-way involved in these 2 drugs induced apoptosis.On the other hand, IFN-a and a-TF were also re-ported to induce cancer cell apoptosis in associa-tion with activation of Bak, Bax, cytochrome-c,and caspase-9,49,50 suggesting that the mitochon-dria-mediated intrinsic pathway also contributesto these 2 agents induced apoptosis. To furtherinvestigate which pathways are involved in the 3-drug–induced apoptosis in SCCHN, we checkedcaspase cascades that existed in both extrinsicand intrinsic pathways in the Tu212 and 886LNcell lines. Western blotting analyses showed thatboth caspase-8 and -9 were activated by all 3drugs as single agents in both cell lines. Theirdownstreammolecules, caspase-3 and PARP, werealso activated in either single agents or 2- to 3-drug combinations. Caspase-3 activity analysisdemonstrated that the 3-drug combination fur-ther increased caspase-3 activity compared to ei-ther single agents or 2-drug combinations. Ourresults imply that both the extrinsic and intrinsicapoptotic pathways are involved in the 3-drugcombination induced apoptosis in SCCHN. A coop-erative effect was thus achieved in the 3-druginduced apoptosis.

In summary, this in vitro study has provided anovel cellular and molecular basis for the com-bined treatment of 13-cRA, IFN-a2a, and a-TF onSCCHN. Cell cycle arrest and increased apoptosismay serve as mechanistic bases for the anti-tumoreffects of the combination therapy. Still furtherinvestigation is essential to clarify the major mo-lecular mediators in triggering both cell cyclearrest and apoptosis by the 3-drug combination indetails. These studies will provide a strong foun-dation for the development of more effective ther-apeutic approaches in the primary or adjuvanttherapeutic setting for SCCHN.

Acknowledgment. We thank Dr. MouradTighiouart for his consultation in statistical anal-ysis.

REFERENCES

1. Hong WK, Endicott J, Itri LM, et al. 13-cis-retinoic acid inthe treatment of oral leukoplakia. N Engl J Med 1986;315:1501–1505.

2. Lotan R. Retinoids in cancer chemoprevention. FASEB J1996;10:1031–1039.

3. Borden EC. Interferons: expending therapeutic roles.N Engl J Med 1992;149:1491–1493.

4. Ohashi M, Yoshida K, Kushida M, et al. Adenovirus-mediated interferon-a gene transfer induces regional di-rect cytotoxicity and possible systemic immunity againstpancreatic cancer. Br J Cancer 2005;93:441–449.

5. Indraccolo S, Tisato V, Tosello V, et al. Interferon-a genetherapy by lentiviral vectors contrasts ovarian cancergrowth through angiogenesis inhibition. Hum Gene Ther2005;16:957–970.

6. Lancillotti F, Giandomenico V, Affabris E, Fiorucci G,Romeo G, Rossi GB. Interferon a-2b and retinoic acidcombined treatment affects proliferation and gene ex-pression of human cervical carcinoma cells. Cancer Res1995;55:3158–3164.

7. Zou C, Ramakumar S, Qian L, et al. Effect of retinoicacid and interferon a-2a on transitional cell carcinoma ofbladder. J Urol 2005;173:247–251.

8. Lingen MW, Polverini PJ, Bouck NP. Retinoic acid andinterferon a act synergistically as antiangiogenic andantitumor agents against human head and neck squa-mous cell carcinoma. Cancer Res 1998;58:5551–5558.

9. Lippman SM, Parkinson DR, Itri LM, et al. 13-cis-reti-noic acid and interferon a-2a: effective combination ther-apy for advanced squamous cell carcinoma of the skin.J Natl Cancer Inst 1992;84:235–241.

10. Lippman SM, Glisson BS, Kavanagh JJ, et al. Retinoicacid and interferon combination studies in human can-cer. Eur J Cancer 1993;29(Suppl. 5):S9–S13.

11. Dimery IW, Hong WK, Lee JJ, et al. Phase I trial of a-to-copherol effects on13-cis-retinoic acid toxicity. Ann Oncol1997;8:85–89.

12. Shin DM, Mao L, Papadimitrakopoulou VM, et al. Bio-chemopreventive therapy for patients with premalignantlesions of head and neck and p53 gene expression. J NatlCancer Inst 2000;92:69–73.

13. Papadimitrakopoulou VM, Liu DD, Mao L, et al. Biologiccorrelates of a biochemoprevention trail in advancedupper aerodigestive tract premalignant lesions. CancerEpidemiol Miomarkers Prev 2002;11:1605–1610.

14. Shin DM, Khuri FR, Murphy B, et al. Combined interferon-a, 13-cis-retinoic acid, and a-tocopherol in locally advancedhead and neck squamous cell carcinoma: novel bioadjuvantphase II trial. J Clin Oncol 2001;19:3010–3017.

15. Seixas-Silva JA Jr, Richards T, Khuri FR, et al. Phase 2bioadjuvant study of interferon a-2a, isotretinoin, andvitamin E in locally advanced squamous cell carcinomaof the head and neck: long-term follow-up. Arch Otolar-yngol Head Neck Surg 2005;131:304–307.

16. Chambon P. The molecular and genetic dissection of theretinoid signaling pathway. Recent Prog Horm Res 1995;50:317–332.

17. Gudas LJ, Sporn MB, Roberts AB. Cellular biology and bio-chemistry of the retinoids. In: Sporn MB, Roberts AB,Goodman DS, editors. The retinoids: biology, chemistryand medicine, 2nd ed. New York: Raven Press; 1994.pp 443–520.


18. Petkovich M. Regulation of gene expression by vitaminA: the role of nuclear retinoic acid receptors. Annu RevNutr 1992;12:443–471.

19. Darnell JE, Kerr IM, Stark GR. Jak-STAT pathways andtranscriptional activation in response to IFNs and otherextracellular signaling proteins. Science 1994;264:1415–1421.

20. Tanaka N, Kawakami T, Taniguchi T. Recognition DNAsequences of interferon regulator 1 (IRF-1) and IRF-2,regulators of cell growth and the interferon system. MolCell Biol 1993;13:4531–4538.

21. Miyamoto M, Fujita T, Kimura Y, et al. Regulated ex-pression of a gene encoding a nuclear factor, IRF-1, thatspecifically binds to IFN-b gene regulatory elements. Cell1988;54:903–913.

22. Skehan P, Storeng R, Scudiero D, et al. New colorimetriccytotoxicity assay for anticancer-drug screening. J NatlCancer Inst 1990;82:1107–1112.

23. Pinto H, Li Y, Loprinzi C, et al. Phase III trail of lowdose 13-cis-retinoic acid for prevention of second primarycancers in stage I–II head and neck cancer—an EasternCooperative Oncology Group Study. Proc Am Soc ClinOncol 2001;20:222a. (abstract No. 886).

24. Khuri F, Lee JJ, Lippman SM, et al. Randomized phaseIII trial of low-dose isotretinoin for prevention of secondprimary tumors in stage I and II head and neck cancerpatients. J Natl Cancer Inst 2006;98:441–450.

25. Sacks PG, Oke V, Amos B, Vasey T, Lotan R. Modulationof growth, differentiation and glycoprotein synthesis byb-all-trans retinoic acid in a multicellular tumor sphe-roid model for squamous carcinoma of the head andneck. Int J Cancer 1989;44:926–933.

26. Sacks PG, Oke V, Vasey T, Lotan R. Retinoic acid inhibi-tion of a head and neck multicellular tumor spheroidmodel. Head Neck 1989;11:219–225.

27. Sacks PG, Racz T, Schantz SP, Rosenblum MG. Growthinhibition by interferon b and g of MDA 886Ln mono-layer cells and multicellular tumor spheroids. A differen-tiation therapy model for squamous cell carcinoma. ArchOtolaryngol Head Neck Surg 1994;120:1267–1272.

28. Lippman SM, Benner SE, Hong WK. Cancer chemopre-vention. J Clin Oncol 1994;12:851–873.

29. Neuzil J, Zhao M, Ostermann G, et al. a-Tocopherol succi-nate, an agent with in vivo anti-tumor activity, induces apo-ptosis by causing lysosomal instability. Biochem J 2002;362:709–715.

30. Ni J, Cheng M, Zhang Y, Li R, Huang J, Yeh S. VitaminE succinate inhibits human prostate cancer cell growthvia modulating cell cycle regulatory machinery. BiochemBiophys Res Commun 2003;300:357–363.

31. Rose AT, McFadden DW. a-Tocopherol succinate inhibitsgrowth of gastric cancer cells in vitro. J Surg Res 2001;95:19–22.

32. Formelli F, Cavadini E, Mascheroni L, Belli F, CascinelliN. Pharmacokinetics and effects on plasma retinol con-centrations of 13-cis-retinoic acid in melanoma patients.Br J Cancer 1997;76:1655–1660.

33. Wills RJ. Clinical pharmacokinetics of interferons. ClinPharmacokinet 1990;19:390–399.

34. Dimitrov NV, Meyer-Leece C, McMillan J, Gilliland D,Perloff M, Malone W. Plasma a-tocopherol concentrations

after supplementation with water- and fat-soluble vitaminE. Am J Clin Nutr 1996;64:329–335.

35. Giannini F, Maestro R, Vukosavljevic T, Pomponi F,Boiocchi M. All-trans, 13-cis and 9-cis retinoic acidsinduce a fully reversible growth inhibition in HNSCCcell lines: implications for in vivo retinoic acid use. IntJ Cancer 1997;70:194–200.

36. Fujimura S, Suzumiya J, Anzai K, et al. Retinoic acidsinduce growth inhibition and apoptosis in adult T-cellleukemia (ATL) cell lines. Leuk Res 1998;22:611–618.

37. Grebenova D, Kuzelova K, Fuchs O, et al. Interferon-asuppresses proliferation of chronic myelogenous leukemiacells K562 by extending cell cycle S-phase without inducingapoptosis. Blood Cells Mol Dis 2004;32:262–269.

38. Sangfelt O, Erickson S, Castro J, et al. Molecular mecha-nisms underlying interferon-a-induced G0/G1 arrest.CKI-mediated regulation of G1 Cdk-complexs and activa-tion of pocket proteins. Oncogene 1999;18:2798–2810.

39. Venkateswaran V, Fleshner NE, Klotz LH. Modulation ofcell proliferation and cell cycle regulators by vitamin Ein human prostate carcinoma cell lines. J Urol 2002;168:1578–1582.

40. Porta C, Hadj-Slimane R, Nejmeddine M, et al. Inter-feron a and g induce p53-dependent and p53-independ-ent apoptosis respectively. Oncogene 2005;24:605–615.

41. Neuzil J, Kagedal K, Andera L, Weber C, Brunk UT.Vitamin E analogs: a new class of multiple action agentswith anti-neoplastic and anti-atherogenic activity. Apo-ptosis 2002;7:179–187.

42. Sun SY, Hail N Jr, Lotan R. Apoptosis as a novel target for can-cer chemoprevention. J Natl Cancer Inst 2004;96:662–672.

43. Hajra KM, Liu JR. Apoptosome dysfunction in humancancer. Apoptosis 2004;9:691–704.

44. Rigobello MP, Scutari G, Friso A, Barzon E, Artusi S,Bindoli A. Mitochondrial permeability transition and re-lease of cytochrome c induced by retinoic acids. BiochemPharmacol 1999;58:665–670.

45. Li C, Chi S, He N, et al. INFa induces Fas expression andapoptosis in hedgehog pathway activated BCC cells throughinhibiting Ras-Erk signaling. Oncogene 2004;23:1608–1617.

46. Liedtke C, Groger N, Manns MP, Trautwein C. Inter-feron-a enhances TRAIL-mediated apoptosis by up-regu-lating caspase-8 transcription in human hepatoma cells.J Hepatol 2006;44:342–349.

47. Shun MC, Yu W, Gapor A, et al. Pro-apoptotic mecha-nisms of action of a novel vitamin E analog (a-TEA) and anaturally occurring form of vitamin E (d-tocotrienol) inMDA-MB-435 human breast cancer cells. Nutr Cancer 2004;48:95–105.

48. Tomasetti M, Rippo MR, Alleva R, et al. a-Tocopherylsuccinate and TRAIL selectively synergise in inductionof apoptosis in human malignant mesothelioma cells. BrJ Cancer 2004;90:1644–1653.

49. Panaretakis T, Pokrovskaja K, Shoshan MC, Grander D.Interferon-a-induced apoptosis in U266 cells is associatedwith activation of the proapoptosis Bcl-2 family membersBak and Bax. Oncogene 2003;22:4543–4556.

50. Yu W, Sanders BG, Kline K. RRR-a-tocopherol succinate-induced apoptosis of human breast cancer cells involvesBax translocation to Mitochondria. Cancer Res 2003;63:2483–2491.


Documents

Induction of cell cycle arrest and apoptosis by a combined treatment with 13-cis-retinoic acid, interferon-α2a, and α-tocopherol in squamous cell carcinoma of the head and neck