cancer research induction of human epithelial stem ... › content › canres › 70 › 22 ›...

Tum

InduExp

EmilioAhma

Abst

Intro

FOXnecessof timviousG2-Msuccesnancefirst ecer, din humgene afactorcomm

AuthorLondonfor ClinKingdoMedicin

Note:Resear

Correstic Oralry, QueStreet,Fax: 44

doi: 10

©2010

www.a

Dow

Published OnlineFirst November 9, 2010; DOI: 10.1158/0008-5472.CAN-10-2173

Dow


Dow


Canceresearch

or and Stem Cell Biology

ction of Human Epithelial Stem/Progenitor

R

ansion by FOXM1

s Gemenetzidis1, Daniela Elena-Costea2, Eric K. Parkinson1,
d Waseem1, Hong Wan1, and Muy-Teck Teh1
ractStem

quent,suggesnot becells ocausedtissuea hyptranscfor FOFOXMthe st

GLI1 (4).on differe

s' AffiliationSchool ofical and Dm and 2See and Odo

Supplemench Online (

ponding AuSciences, Ben Mary UnLondon E1-20-7882-7

.1158/0008-

American A

acrjourna

nloadednloadednloaded

cells are permanent residents of tissues and thought to be targets of cancer initiation. The fre-and often early, upregulation of the FOXM1 transcription factor in the majority of human cancersts that it may participate in the initiation of human tumorigenesis. However, this hypothesis hasen tested. Herein, we show that targeting the ectopic expression of FOXM1 to the highly clonogenicf primary human keratinocytes with stem/progenitor cell properties, but not to differentiating cells,clonal expansion in vitro. We show, using a functional three-dimensional organotypic epithelial

regeneration system, that ectopic FOXM1 expression perturbed epithelial differentiation generatingerproliferative phenotype reminiscent of that seen in human epithelial hyperplasia. Furthermore,riptional expression analysis of a panel of 28 epithelial differentiation-specific genes reveals a roleXM1 in the suppression of epithelial differentiation. This study provides the first evidence that1 participates in an early oncogenic pathway that predisposes cells to tumorigenesis by expandingem/progenitor compartment and deregulating subsequent keratinocyte terminal differentiation. Thisg reveals an important window of susceptibility to oncogenic signals in epithelial stem/progenitor
findin
cells prior to differentiation, and may provide a significant benefit to the design of cancer therapeuticinterventions that target oncogenesis at its earliest incipient stage. Cancer Res; 70(22); 9515–26. ©2010 AACR.

man clationhumakidneyin tumFOXMhumaFOXMfoundFOXMcancein botdepen

duction

M1 directs a transcriptional network of genes that isary for cell cycle promotion and for the coordinationely cell division and exit from the cell cycle (1, 2). Pre-studies have shown its role as a major regulator of thephase of the cell cycle where it is required to ensuresful execution of the mitotic program and the mainte-of chromosome stability (3). In 2002, we provided thevidence that directly linked FOXM1 with human can-emonstrating that FOXM1 was abundantly expressedan basal cell carcinomas and further identified thiss a downstream target of the oncogenic transcription

FOXM1 has since been listed amid the mostntially expressed genes in the majority of hu-

cyclinepitheexecutFOXMmicestrongigeniccyclecontriat all,Squ

manenstasisslow-ccells) istem c

s: 1Queen Mary University of London, Barts and TheMedicine and Dentistry, Institute of Dentistry, Centreiagnostic Oral Sciences, London, England, Unitedction of Pathology, The Gade Institute, Faculty ofntology, University of Bergen, Bergen, Norway

tary data for this article are available at Cancerhttp://cancerres.aacrjournals.org/).

thor: Muy-Teck Teh, Centre for Clinical and Diagnos-arts and The London School of Medicine and Dentist-iversity of London, The Blizzard Building, 4 Newark2AT, United Kingdom. Phone: 44-20-7882-7140;

137; E-mail: [email protected].

5472.CAN-10-2173

ssociation for Cancer Research.

ls.org

on June 1, 2020. © 2cancerres.aacrjournals.org from on June 1, 2020. © 2cancerres.aacrjournals.org from on June 1, 2020. © 2cancerres.aacrjournals.org from

ancers (1, 2). Our finding that transcriptional upregu-of FOXM1 precedes malignancy in a number of solidn cancer types including oral, esophageal, lung, breast,, bladder, and uterine cancers suggests a role for FOXM1or initiation (5). We have also found that ectopic1 expression induces genomic instability in primaryn oral and skin cells (5, 6), which supports a role for1 in both malignant progression and metastatic invasionin a number of human cancer types (1, 2). The fact that1 is implicated throughout early and late stages ofr progression suggests a fundamental role for FOXM1h tumor initiation and maintenance. FOXM1 activity isdent on oncogenic signaling downstream of Ras (7) andD1 (2), the deregulation of which is often implicated inlial tumor initiation and promotion. As an importantor of oncogenic stimulation, the aberrant activation of1 in models of chemically induced carcinogenesis inis tumor-promoting (2, 8, 9), whereas its ablation hastumor-suppressive effects (10, 11). Although the tumor-role for FOXM1 has been attributed to aberrant cellregulation, the mechanism of how its upregulationbutes to cancer initiation in human epithelial cells, ifremains unclear.amous epithelia are continuously maintained by per-t stem cell populations that are responsible for homeo-(12–14). During wound repair, however, a different,ycling population of cells (regenerative stem/progenitor
s responsible for tissue regeneration (15), and much likeells, it is also capable of repopulating all the lineages of
9515

010 American Association for Cancer Research.010 American Association for Cancer Research.010 American Association for Cancer Research.

http://cancerres.aacrjournals.org/



the tisbehavwhencells ato isoAlthouture, tof eithand thClonel

differethat hwordstermerenewonly fodifferseparathe bafacilitalationsexpreconstiretainintegrhand,pool bCloneadenopressoin vivwhenlayerregresby difThe

in restivatiopansireplenin humrent “maint

initiatthe astem/iting emaligdate,beenimmoinsighFOXMgrounof FOdelineto invproto-of itstinocyte stem/progenitor cells as they represent the maintarget

Mate

ClinicThe

Reseaent stmaryanalysdonatsent a

Cell cPrim

lishedand won miwhichpriortumorbeen(5, 33)

Figurethe tipsFOXM1p75NTRvaluesE, endowas bloFACS-sexpandthe quakeratinop75NTRarea ofindividuthat em

Gemenetzidis et al.

Cance9516

Dow


sue in which it resides. Both populations, however,e like clonogenic (Clonehi) populations of keratinocytesexpanded in vitro (12, 13). We therefore refer to theses stem/progenitor as there are currently no methodslate bona fide stem cells from human tissues (12–14).gh progenitor cells can self-renew very efficiently in cul-hey inevitably exit this compartment as a consequenceer differentiation (16, 17), senescence (17–20), or both,en resemble transit-amplifying cells (or paraclones;o) in that they quickly exit the cell cycle and terminallyntiate (18). The highly self-renewing cells form coloniesave a high cloning efficiency on transfer or, in other, have a high capacity for self-renewal and have beend holoclones (18). The progressive reduction in self-al capacity leads to the formation of paraclones, whichrm small or abortive colonies that quickly terminallyentiate. Numerous approaches have been used tote Clonehi from Clonelo keratinocytes (17, 21–23) onsis of cell surface expression signatures. This hasted the investigation into the response of such popu-to various oncogenic stresses. In response to over-

ssion of mitogen-activated protein kinase (24) ortutively active β-catenin (25), Clonehi keratinocytestheir high clonogenic capacity when detached fromin or expand in numbers, respectively. On the otherthe overexpression of c-MYC depletes the Clonehi

oth in vitro (26) and in vivo (27). Furthermore, thelo population can regain its clonogenic capacity byvirus 5E1A, which inactivates a variety of tumor sup-r pathways (28), but not by a single oncogene. In theo setting, the RAS oncogene could induce neoplasiatargeted to the stem/progenitor-containing basalof the epidermis (29), although it can only producesion-prone papillomas when its expression is drivenferentiation-specific promoters (29–31).upregulation of FOXM1 during tissue regeneration

ponse to injury (32) suggests that its physiologic ac-n represents a signal which is required for the ex-on of regenerative stem/progenitor cells for theishment of the epithelium. FOXM1 is an early eventan epithelial neoplasia (5), and in line with the cur-
cancer stem cell” concept that malignancies are clono
al primary human oral keratinocyte strains; bars, SEM. J, graphical representatioerged 15 d after transduction. A total of 200 to 500 colonies were counted (*, P

r Res; 70(22) November 15, 2010

on June 1, 2020. © 2cancerres.aacrjournals.org nloaded from

ion and self-renewal capacity, we hypothesized thatcquisition of aberrant FOXM1 gene expression inprogenitor cells may give rise to progeny cells inher-xcessive FOXM1 levels, as often found in many pre-nant tissues (5) and malignant cancers (1, 2). Tostudies of FOXM1 as a cancer-causing gene haveperformed in the context of transformed and/orrtalized cell lines, undoubtedly providing invaluablet into the molecular mechanisms mediated by1. However, due to the aberrant molecular back-d already present in these cell line systems, the roleXM1 in the earliest stages of neoplasia could not beated and is not fully understood. Herein, we soughtestigate the role of FOXM1 (isoform B; FOXM1B)oncogene, by examining its regulation and the effectoverexpression, in normal primary human oral kera-

s of human malignancies.

rials and Methods

al samplesuse of human tissue was approved by the relevant

rch Ethics Committees at each institution. The pres-udy involved three oral SCC tumor tissue explants pri-cultures for flow cytometry and gene expressionis, and six independent normal oral mucosa tissuesed by healthy volunteers with informed patient con-nd ethical approval.

ultureary normal human oral keratinocytes were estab-from clinically normal oral buccal or gingival tissueere maintained in RM+ medium (5) and were growntomycin (10 μg/mL for 4 h) treated 3T3-feeder layerswere plated at a density of 1.8 × 104/cm2 for 24 hoursto keratinocyte seeding. All established cell lines andexplant culture procedures used in this study have

characterized and cultured as described previously. Retroviral transduction cell synchronization and cell
genic assay protocols are described in the Supple-
ained by a subpopulation of cells that possess tumor mental Data file.

1. A, immunohistochemical staining of p75NTR (green) and FOXM1 (red) in normal human oral mucosa epithelium. B and C, magnified views ofof connective tissue papilla and deep-rete ridges, respectively (dotted boxes) in A are shown here where coexpression (yellow) of p75NTR andare found predominantly in the suprabasal layer and also in isolated cells within the basal cell layer. D, absolute qPCR analysis of FOXM1B andendogenous mRNA expression levels in p75NTR-sorted primary oral keratinocytes 3 d after flow sorting. RS, random sorted populations. All

are representative of three independent experiments. Columns, mean of triplicate samples; bars, SEM (*, P ≤ 0.05; **, P ≤ 0.005; ***, P ≤ 0.001).genous FOXM1 and ΔNp63α protein expression patterns in p75NTR-sorted primary oral keratinocytes following 3 d (i) or 12 d (ii) in culture. β-Tubulintted for protein loading control. F, clonogenicity assays for p75NTR-sorted primary oral keratinocytes. Primary human oral keratinocytes wereorted according to p75NTR levels (low or high) and were plated at equal densities in six-well plates (n = 6). Cells were then allowed to clonallyfor 12 d, after which time they were either trypsinized to obtain cell number values, or were stained with rhodamine B for colony visualization andntification of colony growth. G, p75NTRhi cells were transduced with either (i, ii) pSIN-EGFP or (iii, iv) pSIN-EGFP-FOXM1B and individualcyte colonies were examined under bright and fluorescence microscopy following 3T3-feeder removal. H, clonogenic assays performed onlo and p75NTRhi oral keratinocytes following transduction with either pSIN-EGFP or pSIN-EGFP-FOXM1B. I, quantification of the average colonyp75NTRlo and p75NTRhi oral keratinocytes transduced with either construct. Columns, mean obtained from n = 9 replicates performed in two

2
n of the colony size distribution according to surface area (mm )< 0.05; **, P < 0.01; ***, P < 0.001).
Cancer Research

010 American Association for Cancer Research.


Role for FOXM1 in Human Stem/Progenitor Cell

Cancer Res; 70(22) November 15, 2010www.aacrjournals.org 9517

on June 1, 2020. © 2010 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from



FluorFluo

on BDwashebuffercin) foon icebody/anti-psciencBD Biotwicenally rof 4′,6Integrsortedrespecfrom tthen pwell pmitomserum

OrganOrg

scribeHarrisstitutewere cof Cellgen m

ImmuFor

er laye(Life Tcells paratioformelisted

ImmuFro

acetoPBS) f10% gat roo1% boroomtwice1 houtwice(Thermconcesectiothen ddehyd5 min

unmatempe(pH 9in TBS10 mindescriperfor

Real-ttransAbs

tion-Ptablish(5). FO(YAP1idatedcler 4seque

Resu

Of tfoundHoweinvolvboth n(N/TEin whitiallyarrestMethositionprogenmechation dTo

sion pAs thetiate bwheneSeconits ecttinocya well[a lowknowagreeregionpapillatensepressepibadetecttein i(whichin agr

Gemenetzidis et al.

Cance9518

Dow


escence-activated cell sortingrescence-activated cell sorting (FACS) was performedFACSAria Cell-Sorter (BD Biosciences). Cells wered once with cold PBS and resuspended in cold FACS(PBS/5% fetal bovine serum, 1% penicillin/streptomy-llowed by direct immunofluorescence for 15 minuteswith FITC-conjugated anti-integrin β1 (10 μL anti-

106 cells, CD29-FITC; Abcam), phycoerythrin conjugated75NTR (20 μL antibody/106 cells, CD271-PE; BD Bio-es), or anti-CD44 (1 μL antibody/106 cells, clone G44.26;sciences) in 100 μL FACS buffer. Cells were centrifuged(800 rpm, 3 min) with one PBS wash in-between and fi-esuspended in cold FACS buffer containing 200 ng/mL-diamidino-2-phenylindole (DAPI) dilactate (Sigma).in β1, p75NTR, or CD44 high/low (hi/lo) fractions wereby selecting the highest/lowest 20% of stained cells,tively. Random sorted controls were selected randomlyhe total pool of DAPI-negative cells. Keratinocytes werelated through the automated cell dispenser onto six-lates at 50 to 1,000 cells/cm2, containing preplatedycin C–treated 3T3-feeder layer (1.8 × 104/cm2) or in-free medium.

otypic culture on de-epidermalized dermisanotypic keratinocyte cultures were performed as de-d (22). Tissue processing was carried out by Dr. Wesleyon at the Centre for Cutaneous Research at Blizard In-of Cell and Molecular Sciences, and tissue sectionsut at the Pathology Core Facility of Blizard Instituteand Molecular Sciences. Organotypic culture on colla-atrix protocol is described in Supplementary Data file.

noblottingprotein extraction from keratinocytes growing on feed-rs, cells were incubated for 5 minutes in PBS/Verseneechnologies, Invitrogen; 1:1 ratio) to remove feederrior to protein extraction. Protein extraction and sep-n on SDS-PAGE gels and immunoblotting was per-d as previously described (5). All antibodies used arein the Supplementary Data file.

nofluorescence and confocal microscopyzen tissue sections were fixed in ice-cold methanol/ne (1:1) for 10 minutes or in 4% formaldehyde (inor 20 minutes at room temperature and blocked byoat serum (Sigma) in 0.025% Tween 20 (TBS) for 1 hourm temperature. Incubation with primary antibody (invine serum albumin/TBS) was performed for 2 hours attemperature or overnight at 4°C. Sections were washedin TBS prior to incubation with secondary antibody forr at room temperature. Finally, sections were washedin TBS and mounted with Shandon Immu-Mounto Scientific) solution containing DAPI dilactate (final

ntration at 300 ng/mL). Paraffin-embedded tissuens were initially heated at 60°C for 8 minutes and wereeparaffinized in xylene (three times for 5 min) and
rated (95% ethanol, 90% ethanol, and 70% ethanol foreach) prior to rehydration in water (5 min). Antigen
questeulatio

r Res; 70(22) November 15, 2010


sking was performed by incubation in sub-boilingrature for 20 minutes in antigen retrieval solution.0; DakoCytomation). The sections were washed twiceand permeabilized with TBS 0.2% Triton X (Sigma) forutes at room temperature and were then processed asbed above. Confocal and fluorescence microscopy weremed as described previously (22).

ime absolute quantitative reversecription-PCRolute real-time absolute quantitative reverse transcrip-CR (qPCR) were performed according to previously es-ed protocols involving standard curves for each geneXM1 isoform-specific primers and reference genesand POLR2A) have been previously described and val-(5). Absolute qPCR were performed using the LightCy-80 qPCR system (Roche Diagnostics, Ltd.). Primernces are listed in Supplementary Table S1.

lts and Discussion

he three alternatively spliced isoforms, FOXM1B wasto be the key isoform involved in oncogenesis (1, 2).ver, it remained unclear which isoform(s) was directlyed in the cell cycle. We addressed this question usingormal telomerase-immortalized human keratinocytesRT) and metastatic cervical carcinoma cells (HeLa),ch FOXM1B was the only isoform found to be differen-expressed in cycling cells upon release from growth(Supplementary Fig. S1A and B; see Materials andds). Herein, we investigated the possibility that acqui-of aberrant FOXM1B expression in clonogenic stem/itor cells might be a fundamental oncogenic initiationnism which contributes to its widespread upregula-etected in most human cancers.test this hypothesis, we first investigated the expres-rofile of FOXM1 protein in human oral mucosa tissue.antibody for FOXM1 protein was unable to differen-etween isoforms, we henceforth use the term FOXM1ver isoform-specific information was not available.d, we examined the cellular and molecular effects ofopic expression in primary human oral mucosa kera-te stem/progenitor cells identified and isolated using-established oral stem/progenitor cell marker p75NTR-affinity nerve growth factor receptor (NGFR), alson as CD271, TNFRSF16, Gp80-LNGFR (ref. 23)]. Inment, p75NTR expression was strictly basal, withs of intense positivity at the tips of connective tissuee (Fig. 1A and B), whereas its expression was less in-in deep-rete ridges (Fig. 1C). In contrast, FOXM1 ex-ion was found mainly in the more proliferativesal layers of the oral epithelium and occasionallyed in isolated basal cells (Fig. 1A and C). FOXM1 pro-s mainly cytoplasmic within the normal mucosagenerally has very low proliferative index), this being

eement with the finding that inactive proteins are se-
red in the cytoplasm and that upon mitogenic stim-n, FOXM1 protein translocates into the nucleus (7).
Cancer Research



We hanocytecytoplcytopl

detectthe FO

Figureof FOXhyperpusing thorganoregenerH&E-stof humfrom prkeratinotransdu(A) or pAdjacenorganowith FO(E andindicatebasal arespectdemarcmembr(KRT16transgluand J),K and LM andstained with DAPI (blue).


www.a

Dow


ve further shown that in normal primary oral kerati-s, FOXM1 proteins were predominantly found in the
asm, whereas in the oral SCC cell line UK1 (33), bothasmic and nuclear FOXM1 proteins were ubiquitously
maligareas

acrjournals.org


ed (Supplemental Fig. S1E), which is consistent withXM1 protein expression pattern found in both pre-

2. CharacterizationM1B-inducedroliferation phenotyperee-dimensionaltypic epithelial tissueation model system.ained organotypic culturesan oral mucosa derivedimary human oralcytes retrovirallyced with either pSIN-EGFPSIN-EGFP-FOXM1B (B).t sections of respectivetypic tissues were stainedXM1 (C and D), Ki67F; arrows and arrowheadsexpression of Ki67 in

nd suprabasal cells,ively; dotted lineates the basementane), cytokeratin 16; G and H),taminase-1 (TGM1; Icytokeratin 13 (KRT13;), and filaggrin (FLG;N). Nuclear DNA was

nant dysplastic and oral SCC tumor tissues (5). Inof intense p75NTR staining at the tips of connective

Cancer Res; 70(22) November 15, 2010 9519



tissue(Fig. 1p75NTpatterentiat(contaclonogly pasand pgenicithougwas foFig. Sshoweover sindicamajorwith ttissuecantly(Fig. 1atinocscriptidetectformspecifFOXMstem/as toin vitrepidermaintnegatidenceit is pdirectoral sFOXMer (Figgive rmaintof FOapparFOXMand co(Fig. 1p75NTcell mspecificlonogsupprFOXMmarkein earFig. S3(KRT4paredfindin

latedcells pment,FOXMprecuin aduto theMan

carcinprogeexpresof humFOXMliferatthe exthis, FrovirapSIN-EpSIN-resultcoloniphologratioslargernouslyEGFPincrea(Fig. 1FOXMweresubseshowewith wulatioever, uenhan(Fig.increaof larg(stem/reduc(ref. 1Fig. 1p75NTintrinsuggecells reliferatFOXMsion bTo

typicEGFPcytesIn supgrown

Gemenetzidis et al.

Cance9520

Dow


papillae, FOXM1 expression was almost undetectableB), suggesting an inverse expression pattern betweenR and FOXM1. To further investigate this expressionn, we have FACS-isolated p75NTRlo (containing differ-ing/nonclonogenic cells or paraclones) and p75NTRhi

ining both stem and early progenitor cells which areenic or are able to form holoclones in vitro) from ear-sage (P1) primary normal human oral keratinocyteserformed gene expression (Fig. 1D and E) and clono-ty experiments (Fig. 1F; Supplementary Fig. S2). Al-h integrin β1, an epidermal stem cell marker (16),und to be coexpressed with p75NTR (Supplementary2A), the integrin β1hi/p75NTRhi double-sorted cellsd only marginal advantage in clonogenic potentialingle-sorted p75NTRhi cells (Supplementary Fig. S2C–F),ting that cells expressing p75NTR alone represent theity of clonogenic stem/progenitor cells. In agreementhe inverse expression pattern found in oral mucosaabove, p75NTRhi cells were found to express signifi-lower levels of FOXM1B mRNA (Fig. 1D) and proteinE, i) compared with p75NTRlo or randomly sorted ker-ytes. Although both FOXM1B and FOXM1C are tran-onally active (32), the reduced mRNA expression wased only for FOXM1B isoform (Fig. 1D), but not iso-C (Supplementary Fig. S1C). Given the proliferation-ic role of FOXM1B, we reasoned that the lower1B level was due to the quiescent nature of epithelialprogenitor cells in vivo and in vitro (17, 34, 35), as wellthe slow-cycling status of p75NTRhi keratinocyteso (36) and in vivo (23). Expression analysis of humanmal keratinocyte stem cells in vitro revealed that theenance of their quiescent state is partly attributed to ave regulator of c-MYC, LRIG1 (37). As there is evi-that FOXM1 is a downstream target of c-MYC (38),ossible that LRIG1-induced suppression of c-MYC in-ly contributed to the observed repression of FOXM1 intem/progenitor cells (p75NTRhi). The upregulation of1 protein in cells immediately above the basal cell lay-. 1A–C) led us to hypothesize that p75NTRhi cells mayise to progeny with high clonogenic potential whenained in prolonged culture (Fig. 1F). Indeed, the roleXM1 in progenitor compartment expansion becameent following prolonged culture (12 d) in which1 was significantly upregulated in p75NTRhi cellsnversely, its expression diminished in p75NTRlo cellsE, ii). In both cultures, at days 3 and 12 (Fig. 1E),Rhi cells retained high levels of another epithelial stemarker ΔNp63α (20), indicating that FOXM1 is inducedcally during the expansion of p75NTRhi cells with highenic potential. In contrast, FOXM1 expression wasessed in the differentiating p75NTRlo cells. Indeed,1 and involucrin (IVL, a keratinocyte differentiationr) expression were found to be inversely correlatedly and late passage p75NTRlo cells (SupplementaryA) while another differentiation marker cytokeratin 4) was found to be suppressed in p75NTRhi cells com-
with p75NTRlo cells (Supplementary Fig. S3B). Thesegs further illustrate that FOXM1 expression is upregu-
phenocontro

r Res; 70(22) November 15, 2010


during the expansion of clonogenic stem/progenitorrior to the onset of epithelial differentiation. In agree-during embryonic development, the expression of1 coincides with the transient proliferation of neuralrsors preceding terminal differentiation (39), whereaslt epithelial tissues in mice, its expression is confinedmost proliferative layers (32).y epithelial malignancies, including oral squamous celloma, are thought to arise from premalignant epithelialnitors (40). To understand why FOXM1 is abundantlysed in epithelial premalignancies (5) and a majorityan cancers (1, 2), we investigated the possibility that1 may be an initiating step that confers ectopic pro-ion in keratinocyte stem/progenitor cells leading topansion of potentially premalignant progenies. To testACS-sorted p75NTRlo and p75NTRhi cells were each ret-lly transduced with either pSIN-MCS (empty vector),GFP control, or pSIN-EGFP-FOXM1B. pSIN-MCS andEGFP–transduced cells gave identical experimentals (data not shown). Individual p75NTRhi keratinocytees expressing FOXM1B showed an altered cellular mor-y in which cells exhibited higher nuclear to cytoplasmicwhereas colonies appearedmore compact and relativelyin size (Fig. 1G, compare i and iii). EGFP was homoge-expressed throughout the colony (Fig. 1G, ii), whereas-FOXM1B protein levels were heterogeneous withsed expression towards the periphery of each colonyG, iv). To establish whether ectopic expression of1B confers any functional effects, clonogenic assaysperformed on primary keratinocytes derived from allts (p75NTRlo/hi +EGFP/+FOXM1B). The p75NTRlo cellsd limited growth capacity regardless of the constructhich they have been transduced, consistent with a pop-n containing differentiating/nonclonogenic cells. How-pregulation of FOXM1B in p75NTRhi cells significantlyced clonal expansion over control p75NTRhi+EGFP cells1H–J). p75NTRhi+FOXM1B cells showed a ∼2.6-foldse (compared with p75NTRhi+EGFP) in the percentagee cell colonies representing self-renewing holoclonesprogenitor cell–derived, >10 mm2), and a concomitanttion in the percentage of smaller/abortive colonies8; differentiating/nonclonogenic cell–derived, <5 mm2;J). Hence, the elevation of FOXM1B expression inRhi, but not in p75NTRlo keratinocytes, increased thesic clonogenic capacity of primary keratinocytes. Thissts that the onset of epithelial differentiation rendersfractory to FOXM1B-induced proliferation. To gain pro-ive advantage, we therefore hypothesized that ectopic1B expression promotes stem/progenitor cell expan-y perturbing the onset of terminal differentiation.address this, we performed three-dimensional organo-epithelial regeneration model systems using EGFP or-FOXM1B–transduced primary human oral keratino-according to our previously established method (22).port, EGFP-FOXM1B–transduced oral keratinocytesin the organotypic cultures showed a hyperproliferative
type by producing thicker cell layers compared withl cultures using EGFP-transduced cells (Fig. 2, compare
Cancer Research



FigurePoints,analyse


www.a

Dow


3. Transcriptional regulation of keratinocyte differentiation genes in primary human oral keratinocytes transduced with varying levels of FOXM1B.
mean from triplicate determinations of six independent primary oral keratinocyte strains; bars, SEM. Second-order polynomial regressions were performed to obtain the R2 coefficient of determination values which indicates the degree of correlation between each gene with FOXM1B.
Cancer Res; 70(22) November 15, 2010acrjournals.org 9521

on June 1, 2020. © 2010 American Association for Cancer Research.cancerres.aacrjournals.org nloaded from


betwelevelstrol EGwere alayersimmuconfinorganobasalFOXMmunoing thwithinnonprexpresthe sta

(Suppgestinprolifeepider(growquiredit hascell exactivato be splemeof FOXin epitsal lay

Figureand CEand 19regressF, relatprimary cells extracted from oral SCC tumor tissues (T1–T3), each performed in triplicate. Inset, histogram shows respective CD44 and CD44cell pop

Gemenetzidis et al.

Cance9522

Dow


en A/B and C/D). FOXM1 protein was detected at lowin the basal and proliferative epithelial layers in the con-FP organotypics (Fig. 2C). In contrast, FOXM1 proteinsbundantly expressed in both the basal and suprabasalof the FOXM1B organotypics (Fig. 2D). As expected, thenoreactivity of the proliferation marker Ki67 was strictlyed to the basal proliferative layer of the control EGFPtypics (Fig. 2E). In contrast, keratinocytes of the supra-layers of FOXM1B organotypics retained ectopic1B expression, which correlated with increased Ki67 im-reactivity (Fig. 2F; Supplemental Fig. S4F and G) indicat-at they remained proliferative despite their locationsthe upper spinous layers which should otherwise hostoliferative differentiated cells. Only a subset of FOXM1-

ulations harvested for qPCR analysis.

sing cells are Ki67-positive and this is consistent withining patterns of FOXM1 and Ki67 in oral mucosa tissue

tweentermi

r Res; 70(22) November 15, 2010


lementary Fig. S4E) and basal cell carcinoma (4), sug-g the involvement of FOXM1 in processes other than cellration alone. Single cell expression analysis of humanmal stem cells has revealed that the ablation of EGFRth factor receptor) by LRIG1 (stem cell marker) is re-for the maintenance of stem cell quiescence, although

also been suggested that loss of Lrig1 might trigger stempansion and epithelial hyperproliferation through EGFRtion (37). In agreement with this model, EGFRwas foundignificantly upregulated in FOXM1B organotypics (Sup-ntary Fig. S4A and B) and this further confirms the roleM1B in progenitor compartment expansion resultinghelial hyperproliferation. The expansion of the supraba-ers suggests that FOXM1B perturbed the balance be-

4. Absolute qPCR analysis of FOXM1B mRNA expression and cornifin (SPRR1B; A), involucrin (IVL; B), desmoglein 3 (DSG3; C), p75NTR (D),P55 (E) across a cell panel consisting of 14 normal primary oral keratinocytes, 6 oral premalignant (SVpgC2a, DOK, OKF6/T, POE9n, D19, and D20),oral SCC cell lines (SCC9, SCC25, UK1, CALH2, SCC15, VB6, SqCC/Y1, CaDe12, 5PT, SCC4, H357, and SVFN1–8). Second-order polynomialion analyses were performed to obtain the R2 coefficient of determination values which indicates the degree of correlation between two genes.ive FOXM1B mRNA expression levels in FACS-sorted CD44lo and CD44hi cells from cell lines (POE9n, 5PT, and CA1) and three independent

lo hi

stem/progenitor cell renewal and the commitment tonal differentiation. We therefore hypothesized that

Cancer Research



FOXMepitheTo

keratinditionshown(TGMof theupregua sustcytokeoral epthat iupregFig. S4approsuprabacquirnostaimarke(FLG;proteiSimilausingTo o

ferenttinocyqPCR2P, 3,23) anmaryFOXMdepengenesare baskin a

has alscells ofor maKeratikeratibut thsimpledicatedifferein epitsion (Fa signmore,cornifmoglelated whumaman omouspatterkeratifoundline pp75NTcell pr(55), tindicaationcharaclated ptheseFOXMgenesepithe

Figurethat enand skistem cedisplayprogenwindow


www.a

Dow


1B-induced hyperproliferation might intricately affectlial differentiation.this end, we investigated the expression profile of cyto-16 (KRT16), which is amarker of hyperproliferative con-

s such as wound healing and its upregulation has beento delay differentiation (41), and transglutaminase-1

1), which is specifically expressed in the suprabasal layersoral epithelium (42). Both markers were significantlylated in FOXM1B organotypics (Fig. 2G–J), indicatingained keratinocyte hyperproliferation. Furthermore,ratin 19 (KRT19), which is a marker for preneoplasticithelia (43) and a target of Gli1 in hair follicle stem cellss upregulated in basal cell carcinoma (44), was alsoulated in our FOXM1B-organotypics (SupplementaryC and D). This expression pattern, along with the in-priate mitotic activity of keratinocytes within the upperasal layers, suggests that FOXM1B-expressing cells haveed a defective terminal differentiation. Indeed, immu-ning with early and late keratinocyte differentiationrs cytokeratin 13 (KRT13; Fig. 2K and L) and filaggrinFig. 2M and N), respectively, showed complete lack ofn expression within FOXM1B-derived organotypics.r results were also obtained from organotypic culturescollagen gels (data not shown).btain a more comprehensive understanding of the dif-iation characteristics of FOXM1B-expressing oral kera-tes, we have measured mRNA levels using absoluteon a panel of 21 classic cytokeratin genes (KRT1, 2E,4, 5, 6A, 7, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, andd 2 mesenchymal-specific genes (VIM and NES) in pri-oral keratinocytes transduced with varying doses of1B in culture (Fig. 3). FOXM1B was found to dose-dently transactivate a number of simple basal keratinincluding KRT7, 8, 18, 15, and 19. Keratins 15 and 19
sal cell–specific markers in stratified epithelia includingnd oral mucosa (45–47). Expression of these keratins
Ourupreg

itor compartment to a subsequent “second” oncogenic hit necessary for malignaof cancer susceptibility in the progenitor cells whereby anticancer therapeutic in

acrjournals.org


o been associated with the clonogenic stem/progenitorf the bulge region in hair follicles that are responsibleintenance and regeneration of the epidermis (48, 49).ns 7, 8, and 18 are simple or mixed epithelia–specificns that are normally not expressed in stratified epitheliaey are induced during carcinogenesis (47). Induction ofepithelial–specific and stem cell–specific keratins in-

s that ectopic FOXM1 expression activated a defectiventiation pathway—a common pathologic feature foundhelial malignancies (50). High levels of FOXM1B expres-ig. 3) were found to activate VIM, which may represental for epithelial-mesenchymal transition (51). Further-genes involved in keratinocyte differentiation, such asin (SPRR1B; Fig. 4A), involucrin (IVL; Fig. 4B), and des-in 3 (DSG3; Fig. 4C), were found to be inversely corre-ith FOXM1B mRNA expression across a large panel of

n oral keratinocytes consisting of 14 primary normal hu-ral keratinocytes, 6 oral premalignant, and 19 oral squa-cell carcinoma cell lines. In agreement with expressionns found in oral mucosa (Fig. 1A–C) and primary oralnocytes (Fig. 1D and E, i), p75NTR expression was alsoto be inversely correlated with FOXM1B across this cellanel (Fig. 4D), supporting a tumor suppressor role forR (52–54). Given that upregulation of p75NTR inhibitsoliferation (54) and regulates epithelial differentiationhe inverse expression between FOXM1B and p75NTRtes the deregulation of cell proliferation and differenti-pathways during malignant progression. A previouslyterized FOXM1B target, CEP55 (5), as expected, corre-ositively with FOXM1B expression (Fig. 4E). Collectively,results indicate that the progressive upregulation of1B and downregulation of keratinocyte differentiationmay be intricately associated not only during normallial renewal, but also during malignant progression.
finding that a defined genetic hit, such as FOXM1
ulation, induces the sustained expansion of normal

5. A schematic model showing the mechanism for FOXM1-induced expansion of epithelial progenitor compartment. We have previously shownvironmental factors such as nicotine (5) and UV light (6) can directly activate endogenous FOXM1 (first oncogenic hit) in primary human oraln keratinocytes, respectively. Based on multiple lines of evidence found in this study, we showed that aberrant upregulation of FOXM1 in epitheliallls produce progenitor cells with enhanced proliferative capacity, which affects terminal differentiation, resulting in a hyperplastic phenotypeing perturbed differentiation characteristics. The deregulated differentiation pathway found in hyperplastic tissues may predispose the “expanded”

nt conversion. We hypothesized that FOXM1 mediates atervention at this stage may prevent cancer initiation.




stem/cer stwith tvestigaabsolufrom oestablprimatumorcell mmoreand Chave fexpresafterFOXMgardlelationsis alsotein woral SCtrary toral mthelialwholegests tmaligntumorthat agenitogiving“stemnFurthegenetic-MYCto leadcancerIn s

FOXMstem/in thesexpanTheseregardprogenunder

epitheoncogduringexpan(Fig. 5type opansioexplaias RAstem/thougmultiptationshownUV ligFOXMtively,pathwprovidepitheFOXMmightnant p

Discl

No p

Ackn

We aand thesupportInstitutgifts ofLeigh (regulati

Grant

ThisGemenethe Insttistry, a

Theof pageaccorda

Refe1. My

pro2. Wi

scr3. Lao

of200

4. TeFOCa

Gemenetzidis et al.

Cance9524

Dow


progenitor cells led us to question whether putative can-em cells express different levels of FOXM1 comparedhe rest of the tumor cell population. To test this, we in-ted the endogenous expression levels of FOXM1 usingte qPCR in cancer stem cells which were FACS-isolatedne premalignant oral dysplastic cell line (POE9n), twoished oral SCC cell lines (5PT and CA1; ref. 33), and threery explants of keratinocytes from independent oral SCCs (T1–T3; Fig. 4F), using an established oral cancer stemarker CD44 (56). We have shown that CD44hi cells wereclonogenic than CD44lo cells in oral SCC cell lines (5PTA1; Supplemental Figs. S1F and G). Interestingly, weound that CD44hi and CD44lo cancer cells retain similarsion levels of FOXM1 in both day 3 (Fig. 4F) and day 12FACS isolation (data not shown), suggesting that1 expression is uniformly expressed in tumor cells re-ss of their lineage hierarchy within the tumor subpopu-. The homogenous FOXM1 expression found in tumorsconsistent with our previous report that FOXM1 pro-as abundantly expressed throughout the tumor mass ofCs in vivo (5). In this study, we have shown that con-o its heterogeneous expression pattern found in normalucosal tissue, which retains an intact program of epi-differentiation, FOXM1 is uniformly expressed in thepopulation of tumor-derived keratinocytes. This sug-hat the perturbation of epithelial differentiation duringancy may further perpetuate FOXM1 expression incells. This is in agreement with our current hypothesistumor may arise from a common premalignant pro-r which has acquired high levels of FOXM1 and laterrise to progeny malignant cells (regardless of theiress”) with a hallmark of elevated FOXM1 expression.rmore, malignant cells are characterized by multiplec hits, many of which (e.g., RAS, loss of p53 function,, p16/Rb pathway inactivation, and EGFR) are knownto constitutive upregulation of FOXM1 regardless ofstem cell status.ummary, this study showed the first evidence that1B has a role in the proliferation of normal epithelialprogenitor cell. The aberrant upregulation of FOXM1Be cellsmay perturb the balance between stem/progenitorsion and commitment to terminal differentiation.findings are consistent with the general consensusing epithelial carcinogenesis whereby a premalignant
itor, possibly one with aberrant activation of FOXM1B, Rece
h MT, Wong ST, Neill GW, Ghali LR, Philpott MP, Quinn AG.XM1 is a downstream target of Gli1 in basal cell carcinomas.ncer Res 2002;62:4773–80.

5. Geeaby4:e

6. Teguker

7. MaMEac

r Res; 70(22) November 15, 2010


lium. We have now shown that to achieve a pro-enic “hit,” FOXM1B upregulation ought to be acquireda “permissive window” during stem/progenitor cell

sion prior to their commitment to differentiation). Although unable to revert the differentiated pheno-f committed keratinocytes, FOXM1B induced the ex-n of the stem/progenitor compartment, which mayn why oncogenic stimulation upstream of FOXM1, suchS, is most efficient when targeted to the immatureprogenitor epithelial compartment (30, 31, 57). Al-h large-scale whole-genome sequencing analysis acrossle human cancer types (58) did not find any DNA mu-s within the FOXM1 gene (59), we have previouslythat environmental factors such as nicotine (5) and

ht (6) could directly stabilize and activate endogenous1B in primary human oral and epidermal cells, respec-whereas others have shown that various oncogenicays lead to the activation of FOXM1B (1, 2). This studyes a novel role for FOXM1B in the initiation of humanlial neoplasia, while our previous finding that ectopic1B expression contributed to genomic instability (5, 6)also explain why this gene is implicated in both malig-rogression and metastatic invasion (1, 2).

osure of Potential Conflicts of Interest

otential conflicts of interest were disclosed.

owledgments

re indebted to Professor Farida Fortune (Institute of Dentistry, BartsLondon School of Medicine and Dentistry) for her encouragement and; Professor Ian Mackenzie (Centre for Cutaneous Research, Blizarde of Cell and Molecular Science) for his critical advice and preciousvarious HNSCC cell lines used in this study; and Professor IreneUniversity of Dundee) for her advice on the proliferation-specificon of keratins.

Support

study was co-funded by the Wellcome Trust (M-T. Teh and E.tzidis), Medical Research Council (M-T. Teh and E. Gemenetzidis),itute of Dentistry, Barts and The London School of Medicine and Den-nd the Norwegian Research Council (grant no. 178601; D. Elena-Costea).costs of publication of this article were defrayed in part by the paymentcharges. This article must therefore be hereby marked advertisement innce with 18 U.S.C. Section 1734 solely to indicate this fact.

ived 06/18/2010; revised 09/02/2010; accepted 09/17/2010; published
goes excessive rounds of cl onal expansion within the OnlineFirst 11/09/2010.
rencesatt SS, Lam EW. The emerging roles of forkhead box (Fox)teins in cancer. Nat Rev Cancer 2007;7:847–59.erstra I, Alves J. FOXM1, a typical proliferation-associated tran-iption factor. Biol Chem 2007;388:1257–74.ukili J, Kooistra MR, Bras A, et al. FoxM1 is required for executionthe mitotic programme and chromosome stability. Nat Cell Biol5;7:126–36.

menetzidis E, Bose A, Riaz AM, et al. FOXM1 upregulation is anrly event in human squamous cell carcinoma and it is enhancednicotine during malignant transformation. PLoS One 2009;4849.h MT, Gemenetzidis E, Chaplin T, Young BD, Philpott MP. Upre-lation of FOXM1 induces genomic instability in human epidermalatinocytes. Mol Cancer 2010;9:45.
RY, Tong TH, Cheung AM, Tsang AC, Leung WY, Yao KM. Raf/K/MAPK signaling stimulates the nuclear translocation and trans-tivating activity of FOXM1c. J Cell Sci 2005;118:795–806.
Cancer Research



8. KatraproCa

9. Katorneg200

10. GutidJ C

11. Watrana

12. Basm100

13. Jakhai

14. Snthe201

15. ItotribMe

16. Johu

17. LihuPro

18. Baent230

19. Ma(inhmaFA

20. Pecyt

21. Jontraand

22. Watiescyt

23. Nakeran62

24. Zhactin v

25. Zhtialad

26. Gaep

27. Arnreseny

28. Bagroon

29. Scma200

30. Balomsup

31. Greep

ic99

32. Yehech16

33. Hapacin

34. Biclab82

35. Mocapig23

36. Okcece

37. Jeepof11

38. BlaForec48

39. Ueceem

40. Huca

41. Paprotio

42. Taglutio63

43. Linsiolium

44. Yoce12

45. WaifieDe

46. LlosqJ C

47. Caprefila

48. Liutat20

49. Mimacenu10

50. Owce

51. Vaexpthe

52. Qu


www.a

Dow


lin TV, Wang IC, Ackerson TJ, et al. Increased levels of the FoxM1nscription factor accelerate development and progression ofstate carcinomas in both TRAMP and LADY transgenic mice.ncer Res 2006;66:1712–20.linichenko VV, Major ML, Wang X, et al. Foxm1b transcription fac-is essential for development of hepatocellular carcinomas and isatively regulated by the p19ARF tumor suppressor. Genes Dev4;18:830–50.sarova GA, Wang IC, Major ML, et al. A cell-penetrating ARF pep-e inhibitor of FoxM1 in mouse hepatocellular carcinoma treatment.lin Invest 2007;117:99–111.ng IC, Meliton L, Ren X, et al. Deletion of Forkhead Box M1nscription factor from respiratory epithelial cells inhibits pulmo-ry tumorigenesis. PLoS One 2009;4:e6609.rker N, van Es JH, Kuipers J, et al. Identification of stem cells inall intestine and colon by marker gene Lgr5. Nature 2007;449:3–7.s V, Barker N, Kasper M, et al. Lgr5 marks cycling, yet long-lived,r follicle stem cells. Nat Genet 2008;40:1291–9.ippert HJ, Haegebarth A, Kasper M, et al. Lgr6 marks stem cells inhair follicle that generate all cell lineages of the skin. Science0;327:1385–9.M, Liu Y, Yang Z, et al. Stem cells in the hair follicle bulge con-ute to wound repair but not to homeostasis of the epidermis. Natd 2005;11:1351–4.nes PH, Harper S, Watt FM. Stem cell patterning and fate inman epidermis. Cell 1995;80:83–93.A, Simmons PJ, Kaur P. Identification and isolation of candidateman keratinocyte stem cells based on cell surface phenotype.c Natl Acad Sci U S A 1998;95:3902–7.rrandon Y, Green H. Three clonal types of keratinocyte with differ-capacities for multiplication. Proc Natl Acad Sci U S A 1987;84:2–6.urelli R, Zambruno G, Guerra L, et al. Inactivation of p16INK4aibitor of cyclin-dependent kinase 4A) immortalizes primary hu-n keratinocytes by maintaining cells in the stem cell compartment.SEB J 2006;20:1516–8.llegrini G, Dellambra E, Golisano O, et al. p63 identifies keratino-e stem cells. Proc Natl Acad Sci U S A 2001;98:3156–61.es PH, Watt FM. Separation of human epidermal stem cells fromnsit amplifying cells on the basis of differences in integrin functionexpression. Cell 1993;73:713–24.

n H, Yuan M, Simpson C, et al. Stem/progenitor cell-like proper-of desmoglein 3dim cells in primary and immortalized keratino-

e lines. Stem Cells 2007;25:1286–97.kamura T, Endo K, Kinoshita S. Identification of human oralatinocyte stem/progenitor cells by neurotrophin receptor p75d the role of neurotrophin/p75 signaling. Stem Cells 2007;25:8–38.u AJ, Haase I, Watt FM. Signaling via β1 integrins and mitogen-ivated protein kinase determines human epidermal stem cell fateitro. Proc Natl Acad Sci U S A 1999;96:6728–33.u AJ, Watt FM. β-Catenin signalling modulates proliferative poten-of human epidermal keratinocytes independently of intercellular

hesion. Development 1999;126:2285–98.ndarillas A, Watt FM. c-Myc promotes differentiation of humanidermal stem cells. Genes Dev 1997;11:2869–82.old I, Watt FM. c-Myc activation in transgenic mouse epidermisults in mobilization of stem cells and differentiation of their prog-. Curr Biol 2001;11:558–68.rrandon Y, Morgan JR, Mulligan RC, Green H. Restoration ofwth potential in paraclones of human keratinocytes by a viralcogene. Proc Natl Acad Sci U S A 1989;86:4102–6.holes AG, Woolgar JA, Boyle MA, et al. Synchronous oral carcino-s: independent or common clonal origin? Cancer Res 1998;58:3–6.illeul B, Surani MA, White S, et al. Skin hyperkeratosis and papil-a formation in transgenic mice expressing a ras oncogene from a
rabasal keratin promoter. Cell 1990;62:697–708.enhalgh DA, Rothnagel JA, Quintanilla MI, et al. Induction of
idermal hyperplasia, hyperkeratosis, and papillomas in transgen-

accre32

acrjournals.org


mice by a targeted v-Ha-ras oncogene. Mol Carcinog 1993;7:–110.H, Kelly TF, Samadani U, et al. Hepatocyte nuclear factor 3/forkad homolog 11 is expressed in proliferating epithelial and mesen-ymal cells of embryonic and adult tissues. Mol Cell Biol 1997;17:26–41.rper LJ, Piper K, Common J, Fortune F, Mackenzie IC. Stem celltterns in cell lines derived from head and neck squamous cell car-oma. J Oral Pathol Med 2007;36:594–603.kenbach JR, Mackenzie IC. Identification and localization ofel-retaining cells in hamster epithelia. J Invest Dermatol 1984;:618–22.rris DJ, Reis A. A YAC contig spanning the nevoid basal cellrcinoma syndrome, Fanconi anaemia group C, and xerodermamentosum group A loci on chromosome 9q. Genomics 1994;:23–9.umura T, Shimada Y, Imamura M, Yasumoto S. Neurotrophin re-ptor p75(NTR) characterizes human esophageal keratinocyte stemlls in vitro. Oncogene 2003;22:4017–26.nsen KB, Watt FM. Single-cell expression profiling of humanidermal stem and transit-amplifying cells: Lrig1 is a regulatorstem cell quiescence. Proc Natl Acad Sci U S A 2006;103:958–63.nco-Bose WE, Murphy MJ, Ehninger A, et al. C-Myc and its targetxM1 are critical downstream effectors of constitutive androstaneeptor (CAR) mediated direct liver hyperplasia. Hepatology 2008;:1302–11.no H, Nakajo N, Watanabe M, Isoda M, Sagata N. FoxM1-drivenll division is required for neuronal differentiation in early Xenopusbryos. Development 2008;135:2023–30.nter KD, Parkinson EK, Harrison PR. Profiling early head and neckncer. Nat Rev Cancer 2005;5:127–35.ladini RD, Coulombe PA. Directed expression of keratin 16 to thegenitor basal cells of transgenic mouse skin delays skin matura-n. J Cell Biol 1998;142:1035–51.BM, Gallagher GT, Chakravarty R, Rice RH. Keratinocyte trans-taminase in human skin and oral mucosa: cytoplasmic localiza-n and uncoupling of differentiation markers. J Cell Sci 1990;95:1–8.dberg K, Rheinwald JG. Suprabasal 40 kd keratin (K19) expres-n as an immunohistologic marker of premalignancy in oral epithe-. Am J Pathol 1989;134:89–98.

ussef KK, Van Keymeulen A, Lapouge G, et al. Identification of thell lineage at the origin of basal cell carcinoma. Nat Cell Biol 2010;:299–305.seem A, Dogan B, Tidman N, et al. Keratin 15 expression in strat-d epithelia: downregulation in activated keratinocytes. J Investrmatol 1999;112:362–9.yd C, Yu QC, Cheng J, et al. The basal keratin network of stratifieduamous epithelia: defining K15 function in the absence of K14.ell Biol 1995;129:1329–44.sanova ML, Bravo A, Jorcano JL. Simple epithelial keratins: ex-ssion, function and disease. In: Paramio J, editor. Intermediatements. Springer, US; 2006, p. 110–9.Y, Lyle S, Yang Z, Cotsarelis G. Keratin 15 promoter targets pu-ive epithelial stem cells in the hair follicle bulge. J Invest Dermatol03;121:963–8.chel M, Torok N, Godbout MJ, et al. Keratin 19 as a biochemicalrker of skin stem cells in vivo and in vitro: keratin 19 expressinglls are differentially localized in function of anatomic sites, and theirmber varies with donor age and culture stage. J Cell Sci 1996;109:17–28.ens DM, Watt FM. Contribution of stem cells and differentiatedlls to epidermal tumors. Nat Rev Cancer 2003;3:444–51.ughan MB, Ramirez RD, Andrews CM, Wright WE, Shay JW. H-rasression in immortalized keratinocytes produces an invasive epi-lium in cultured skin equivalents. PLoS One 2009;4:e7908.ann EJ, Khwaja F, Zavitz KH, Djakiew D. The aryl propionic
id R-flurbiprofen selectively induces p75NTR-dependent de-ased survival of prostate tumor cells. Cancer Res 2007;67:54–62.



53. Khhibthe620

54. Jinthe471

55. NaDjacrotio

56. Pri

sune97

57. BronaH-8:5

58. Bigan

Gemenetzidis et al.

Cance9526

Dow


waja F, Allen J, Lynch J, Andrews P, Djakiew D. Ibuprofen in-its survival of bladder cancer cells by induced expression ofp75NTR tumor suppressor protein. Cancer Res 2004;64:7–13.H, Pan Y, Zhao L, et al. p75 neurotrophin receptor suppressesproliferation of human gastric cancer cells. Neoplasia 2007;9:–8.lbandian A, Pang AL, Rennert OM, Chan WY, Ravindranath N,kiew D. A novel function of differentiation revealed by cDNA mi-
array profiling of p75NTR-regulated gene expression. Differentia-n 2005;73:385–96.nce ME, Sivanandan R, Kaczorowski A, et al. Identification of a
59. FoSomu

r Res; 70(22) November 15, 2010


bpopulation of cells with cancer stem cell properties in head andck squamous cell carcinoma. Proc Natl Acad Sci U S A 2007;104:3–8.wn K, Strathdee D, Bryson S, Lambie W, Balmain A. The malig-nt capacity of skin tumors induced by expression of a mutantras transgene depends on the cell type targeted. Curr Biol 1998;16–24.nell GR, Greenman CD, Davies H, et al. Signatures of mutationd selection in the cancer genome. Nature 2010;463:893–8.rbes SA, Tang G, Bindal N, et al. COSMIC (the Catalogue of
matic Mutations in Cancer): a resource to investigate acquiredtations in human cancer. Nucleic Acids Res 2010;38:D652–7.
Cancer Research



Correction

Correction: Induction of Human Epithelial Stem/Progenitor Expansion by FOXM1

In this article (Cancer Res 2010;70:9515–26), which was published in the November 15,2010 issue of Cancer Research (1), there is a typographical error in the name of thesecond author. The correct name is Daniela Elena Costea.

Reference1. Gemenetzidis E, Costea DE, Parkinson EK, Waseem A, Wan H, Teh M-T. Induction of human

epithelial stem/progenitor expansion by FOXM1. Cancer Res 2010;70:9515–26.

Published online January 3, 2011.�2011 American Association for Cancer Research.doi: 10.1158/0008-5472.CAN-10-4082

CancerResearch

Cancer Res; 71(1) January 1, 2011290

2010;70:9515-9526. Published OnlineFirst November 9, 2010.Cancer Res Emilios Gemenetzidis, Daniela Elena-Costea, Eric K. Parkinson, et al. FOXM1Induction of Human Epithelial Stem/Progenitor Expansion by

Updated version

10.1158/0008-5472.CAN-10-2173doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2010/11/05/0008-5472.CAN-10-2173.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/70/22/9515.full#ref-list-1

This article cites 58 articles, 23 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/70/22/9515.full#related-urls

This article has been cited by 9 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] at

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

Permissions

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

.http://cancerres.aacrjournals.org/content/70/22/9515To request permission to re-use all or part of this article, use this link

on June 1, 2020. © 2010 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from


http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-10-2173

http://cancerres.aacrjournals.org/content/suppl/2010/11/05/0008-5472.CAN-10-2173.DC1

http://cancerres.aacrjournals.org/content/70/22/9515.full#ref-list-1

http://cancerres.aacrjournals.org/content/70/22/9515.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/70/22/9515


cancer research induction of human epithelial stem ... › content › canres › 70 › 22 ›...

Documents