Rab25 Regulates Invasion and Metastasis in Head and Neck ...Advancing our understanding of the molecular mechan- ... may lead to more effective treatment options. Invasion and metastasis

Human Cancer Biology

Rab25 Regulates Invasion and Metastasis in Head andNeck Cancer

Panomwat Amornphimoltham1, Kamil Rechache1, Jamie Thompson1, Andrius Masedunskas1,Kantima Leelahavanichkul2, Vyomesh Patel2, Alfredo Molinolo2, J. Silvio Gutkind2, and Roberto Weigert1

AbstractPurpose:Head andneck squamous cell carcinoma (HNSCC) is one of the 10most common cancers with

a 50% five-year survival rate, which has remained unchanged for the past three decades. One of the major

reasons for the aggressiveness of this cancer is that HNSCCs readilymetastasize to cervical lymph nodes that

are abundant in the head and neck region. Hence, discovering new molecules controlling the metastatic

process as well as understanding their regulation at themolecular level are essential for effective therapeutic

strategies.

Experimental Design: Rab25 expression level was analyzed in HNSCC tissue microarray. We used a

combination of intravital microscopy in live animals and immunofluorescence in an in vitro invasion assay

to study the role of Rab25 in tumor cell migration and invasion.

Results: In this study, we identified the small GTPase Rab25 as a key regulator of HNSCCmetastasis. We

observed that Rab25 is downregulated in HNSCC patients. Next, we determined that reexpression of Rab25

in ametastatic cell line is sufficient to block invasion in a three-dimensional collagenmatrix andmetastasis

to cervical lymph nodes in a mouse model for oral cancer. Specifically, Rab25 affects the organization of

F-actin at the cell surface, rather than cell proliferation, apoptosis, or tumor angiogenesis.

Conclusion: These findings suggest that Rab25 plays an important role in tumor migration and

metastasis, and that understanding its function may lead to the development of new strategies to prevent

metastasis in oral cancer patients. Clin Cancer Res; 19(6); 1375–88. �2013 AACR.

IntroductionThe term head and neck cancer encompasses a series of

cancers that arise from the oral and nasal cavity, salivaryglands, paranasal sinuses, oropharynx, hypopharynx, andlarynx. More than 90% of these cancers are squamous cellcarcinomas (SCC), which originate from the squamousmucosa lining the upper aerodigestive tract (1). Head andneck squamous cell carcinoma (HNSCC) is very aggressiveand often invades the cervical lymph nodes and metasta-sizes to distant sites.Distantmetastasis can adversely impactsurvival and significantly affect treatment planning. Thecombination of conventional surgery or radiotherapy andchemotherapy is still the treatment of choice for HNSCCpatients (2). However, the 5-year survival rate (50%) of

these patients has not improved in the past 3 decades (3).Advancing our understanding of the molecular mechan-isms regulating tumor invasion and metastasis for HNSCCmay lead to more effective treatment options.

Invasion and metastasis are the result of a complexinterplay of multiple processes including: angiogenesis oflymphatic and blood vessels, loss of cell polarity, loss ofcell–cell contact, extracellular matrix breakdown mediatedby specific metalloproteases, remodeling of the actin cyto-skeleton, and tumor proliferation at the secondary sites (4,5). Recent studies have shown that molecules involved inintracellular membrane trafficking steps, endocytosis inparticular, control some of these processes (6, 7). Amongthesemolecules,members of the Rab family are deregulatedin various cancers (8). Rab proteins are considered masterregulators of membrane trafficking. With at least 70 mem-bers in the human genome, they constitute the largestfamily of small GTPases. However, Rabs share limitedhomology to each other (40%) and individual Rab familymembers play distinct biologic roles (9, 10). Rabs regulateboth endocytic and secretory pathways where they controldifferent events including protein sorting, membrane lipidcomposition, organelle biogenesis, interaction with cyto-skeletal elements, and membrane fission (11). Rab mole-cules that regulate the secretory pathway are elevated indifferent cancer types, such as Rab1 in human tongue cancer

Authors' Affiliations: 1Intracellular Membrane Trafficking Unit, and2Molecular Carcinogenesis Unit, Oral and Pharyngeal Cancer Branch,National Institute of Dental and Craniofacial Research, NIH, Bethesda,Maryland

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

Corresponding Author: Roberto Weigert, National Institute of Dental andCraniofacial Research, NIH, 30 Convent Drive, Bethesda, MD 20892-4340.Phone: 301-496-9969; Fax: 301-496-1966; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-12-2858

�2013 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org 1375

on March 23, 2020. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst January 22, 2013; DOI: 10.1158/1078-0432.CCR-12-2858

http://clincancerres.aacrjournals.org/

(12), Rab3 in cancers of the nervous system and neuroen-docrine cells, Rab22b/31 in breast cancer (13), and Rab23in hepatic and gastric cancer (14, 15). Endocytic Rabs arealso deregulated in cancer. For example, Rab5a is elevated inhepatocellular carcinoma and thyroid adenoma (16, 17),Rab5b is decreased in metastatic melanoma, and Rab7 andRab20 are upregulated in thyroid adenoma and pancreaticcarcinoma, respectively (13).

Recently, more attention has been directed towardRab25, a member of the Rab11 subfamily, which is exclu-sively expressed in epithelial cells. Rab25has beenproposedto regulate the recycling of proteins from the endosomes tothe plasma membrane (18). However, the associationbetween Rab25 deregulation and cancer is still a matter ofdebate. For example, Rab25 expression is elevated andcorrelated to poor prognosis in prostate and ovarian cancer.On the other hand, Rab25 has also been reported to be bothup- and downregulated in breast cancer (19–21). Morerecently, loss of Rab25 has been shown to promote intes-tinal neoplasia in mice and to correlate with human colo-rectal adenocarcinomas (22). These evidences indicate animportant role of Rab25 in tumor progression and aggres-siveness. Several mechanisms have been proposed to definethe role of Rab25 in cancer. Cheng and colleagues reportedthat overexpression of Rab25 inhibits apoptosis and autop-hagy by increasing cellular bioenergetics (23). Others havesuggested that Rab25, together with the chloride intracel-lular channel 3 (CLIC3), regulates tumor invasiveness andmediates recycling of a5b1-integrin to the plasma mem-brane from a late endosomal compartment (5). Althoughthe mechanism remains unclear, these pieces of evidencestrongly implicate Rab25 in tumor development, progres-sion, and aggressiveness.

In this study, we investigated the role of Rab25 and itsregulation inHNSCC. First, we determined theRab25 statusin normal oral mucosa and HNSCC tissues of differinggrade and stage. Next, we used a lentiviral expression systemtomodify Rab25 expression in SCCcells to study pathologicfunctions of tumor cells. Finally, we investigated the role ofRab25 in tumor development and progression by using acombination of intravital two-photon microscopy (TPM)and immunohistochemistry (IHC) in a mouse model oforal cancer.

Materials and MethodsCell lines

Human cancer cell lines HN12, HeLa-O3, Cal27,UMCCC2, and UMSCC17B maintained as previouslydescribed (24) in Dulbecco’s modified Eagle’s medium(DMEM) supplemented with 10% FBS, at 37�C in 95%air/5% CO2. ORL48 (25) and ORL150 were kindly giftsfrom Dr. Cheong Sok Ching, CARIF, Malaysia. HeLa-A,SiHa, HaCaT, C33A, and A549 cell lines were obtainedfrom American Type Culture Collection andHeLa-J (kindlygift from Dr. Julie Donaldson, NHLBI, NIH) maintainedaccording to the company instruction. Human immortal-ized normal oral keratinocytes (HNOK) were established asdescribed (26). All above cell lines underwent short tandemrepeat DNA authentication (Genetica DNA Laboratories,Inc.) before thedescribed experiments to ensure consistencyin cell identity.

Plasmids constructsHuman Rab25 and Rab11a full-length cDNA clones

were obtained from Mammalian Gene Collection (MGC,NIH) and subcloned into pENTRsfiI intermediate plasmidby fusing it with venus or red fluorescent protein (RFP)sequence at the C-terminal. The C-terminal hypervariableregions of Rab25 (amino acid 171-213) and Rab11 (aminoacid 170-216) were generated by PCR-cloning and fusedwith the N-terminal regions of Rab11 (amino acid 1-169)and Rab25 (amino acid 1-170), respectively. Lifeact GFPwas a gift from Tamas Balla (NICHD, NIH). Histone-GFPfusion (H2B-GFP) was obtained from Addgene (#11680;ref. 27).

Antibodies and reagentsThe following antibodies were used in this study: rabbit

polyclonal anti-Rab25 and cleaved-caspase-3 (Cell Signal-ing Technology), rabbit polyclonal anti-Rab25 (recognizedamino acid residues 131-145, Sigma -Aldrich), rabbit poly-clonal antisera against EGF receptor (EGFR) and tubulin(Santa Cruz Biotechnology), rabbit polyclonal anti-GFPand Lyve1 (Abcam), rabbit polyclonal anti-Ki67 (NovaCastra, Leica microsystems), rat monoclonal anti-CD31(BD Pharmingen). All antibodies were used for Westernblot analysis or IHC at a dilution of 1:1000 or 1:100,respectively. AlexaFluor 488, 594, and 647 conjugatedsecondary antibodies for immunofluorescence were pur-chased from Invitrogen. EGF, Latrunculin A, and Cytocha-lasin D were obtained from Sigma.

Translational RelevanceIn this study, we identified the small GTPase Rab25 as

a key player in the invasion and metastasis in head andneck squamous cell carcinoma (HNSCC). Prompted bythe observation that Rab25 is downregulated at theprotein level in HNSCC patients, we exploited a recentlydeveloped xenograft model in the tongue of immuno-compromised mice to further elucidate its role inHNSCC. Strikingly, we observed that reexpression ofRab25 is sufficient to block invasion and metastasis tolocoregional lymph nodes. Moreover, by using high-resolution intravital microscopy and immunofluores-cence in a three-dimensionalmodel, we determined thatRab25 expression affects cell migration by altering theactin cytoskeleton arrangement at the cell surface, ratherthan affecting cell proliferation, apoptosis, or tumorangiogenesis.These findings indicate that Rab25 may be a promis-

ing therapeutic target and that unraveling its functionmay lead to the development of new strategies to preventmetastasis in oral cancer patients.

Amornphimoltham et al.

Clin Cancer Res; 19(6) March 15, 2013 Clinical Cancer Research1376




Lentiviral expression systemcDNAs encoding for TagRFP, H2B-GFP, venus, venus-

Rab25, venus-Rab25-11, venus-Rab11-25, RFP-Rab25, andLifeact GFP were subcloned into the intermediate vectorpENTRsfiI and transferred to the lentiviral expression vectorpLESIP (28). Short hairpin RNAs (shRNA) targeting non-silencing scramble sequence (pGIPZ sh-scramble) and 3different sequences of Rab25 (pGIPZ sh-Rab25; clone IDV2LHS_38594, V3LHS_362078, V3LHS_362078) wereobtained from Open Biosystem (Thermo Scientific). Lenti-viral stocks were prepared and titrated with HEK-293T cellsas packaging cells. Tumor cells were infected with the virusfor 16 hours. After, cells were returned to normal growthmedium and infected cells were isolated with fluorescent-activated cell sorting (FACS) and maintained under puro-mycin (1 mg/mL) selection.

Tissue arrays immunohistochemistryOral cavity SCC and normal tissues high-density tissue

microarray with grade and tumor–node–metastasis (TNM;stage; 69 cases/208 cores, #OR 208) were purchased fromUS BioMax Inc. and used for IHC. The formalin-fixedparaffin-embedded tumor tissue array slides were firstdeparaffinized in 100% xylene. The slides were hydratedthrough a series of graded alcohols (100%, 95%, 80%, and70%) for 5 minutes each, washed once in H2O for 5minutes, incubated in sodium citrate buffer (pH6.0) for2 minutes in microwave at full power, and then for 20minutes at 10% power to unmask the antigen. The slideswere then incubated in: (i) 3% hydrogen peroxide at roomtemperature for 10 minutes to quench endogenous perox-idase activity; (ii) blocking serum (2% bovine serum albu-min in PBS-1% Tween 20) for 1 hour; and (iii) primaryantibody in blocking buffer overnight at 4�C. The slideswere washed in PBS 3 times, incubated with a biotin-conjugated secondary antibody (1:400 dilution) at roomtemperature for 30 minutes followed by the avidin-biotincomplex (Vector Stain Elite, ABC kit, Vector Laboratories)for 30minutes at room temperature. The slideswerewashedand developed in 3,30-diaminobenzidine (Sigma FASTDABtablet, Sigma Chemical) under microscopic observation.The reaction was stopped in tap water and the tissues werecounterstained with Mayer’s hematoxylin, dehydrated, andmounted. The images were taken using Scanscope (Aperio).The evaluation of the IHC was conducted blindly byA. Molinolo and P. Amornphimoltham. The whole tissuearray slides were examined and classified based on thestaining intensity (1, weak staining; 2, moderate staining;and 3, strong staining) and the percentage of positive cellsquantified as previously described (0, <10%of stained cells;1, 10%–25%; 2, 25%–50%; 3, 50–75%; and 4, 75–100%ofcells stained; refs. 28, 29).

Three-dimensional collagen-matrix invasion assayCollagen gels (rat tail collagen type I, BDBioscience)were

prepared in serum-free DMEM at a final concentration of2.0 mg/mL and added (500 mL volume) to transwell inserts(0.4 mm polycarbonated membrane) of 12-well plates

(Costar). Collagen was allowed to polymerize and equili-brate in culture incubator for 2 hours. After, 200 mL ofserum-free media containing 0.4 � 106 tumor cells wereadded to each well. Serum-containing, serum-free mediawithout or with 10 ng/mL EGF were added into the bottomwells. When needed, inhibitors of the actin cytoskeletonwere added in the transwell insert at the beginning of assay(1 mmol/L Latrunculin A and 10 mmol/L Cytochalasin D).Cells were allowed to invade in the three-dimensional (3D)collagen gels for 8 hours. Culture media was removed andthe gels containing the tumor cells were fixed in 4% form-aldehyde in PBS for 30 minutes. Z stack images wereobtained using an inverted confocal Olympus microscope(water objective XLUMPFL20XW, numerical aperture (NA)0.95). Fluorescent-labeled polystyrene microspheres (15mm; Invitrogen) were used to mark the top of the gel. Toquantify tumor cells invasion, 8 random fields per condi-tion in at least 2 independent experiments were scannedand the distance of the invading cells from the top of the gelwas measured. Data are reported as mean distance ofinvading cell (�SD).

Tongue tumorxenograft inathymicnu/nuorSCIDmiceAll the experiments were approved by the National Insti-

tute of Dental and Craniofacial Research Animal Care andUse Committee (NIH). Female athymic (nu/nu) nude mice(Harlan Sprague Dawley) and female severe combinedimmunodeficient (SCID) mice (National Cancer Instituteat Frederick) 5–6 weeks old and 20–25 g, were used in thestudy. The mice were housed in appropriate sterile filter-capped cages, fed and watered ad libitum. Half a million ofHN12 or HeLa-O3 cells were submucosally injected in thelateral anterior of the tongue. Animals were fed with softdough diet from the day of injection. The tumor growthwasmonitored on a weekly basis. For each experiment, tumor-bearing animals were randomly divided into groups of 10animals. All experiments were done in triplicate. At theindicated time points, animals were euthanized and tissueswere collected. Tissues were lysed in protein lysis buffer forWestern blot analysis. Alternatively, they were fixed andembedded in paraffin, or frozen and embedded in optimalcutting temperature compound (OCT; Tissue-Tek, SakuraFinetek) for histopathologic assessment and IHC study.

ImmunofluorescenceFor immunofluorescence staining, OCT-embedded fro-

zen tissues were cut (15 mm) and put onto silanated glassslides, air-dried, and stored at �80�C. Cryosections werethawed at room temperature, hydrated, washed with PBS,and incubated in blocking solution (10% heated-inacti-vated FBS in 0.01% saponin PBS) for 1 hour followed byincubation with the first primary antibody diluted in block-ing solutions at 4�C, overnight. After washing, slides weresequentially incubated with the appropriate fluorescent-conjugated secondary antibody (dilution, 1:750) for 30minutes. The slides were washed and incubated for 10minutes in Hoechst 33342 (Invitrogen) to label nuclei, andmounted in Fluoromount G (Southern Biotech). Images

Rab25 in Head and Neck Cancer Metastasis

www.aacrjournals.org Clin Cancer Res; 19(6) March 15, 2013 1377




were acquired and analyzed using a confocalmicroscope, asdescribed below.

Tongue tumor imaging with two-photon and confocalmicroscopy

Mice were anesthetized by intraperitoneal injections ofketamine (10 mg/kg) and xylazine (100 mg/kg). The ani-mals were placed on a preheated stage with themouth openand the tongue gently retracted using nontooth forceps. Thetongue was hold in a custom-made holder, secured with aglass coverslip (Fig. 4A). The preheated stage was movedclose to the objective of a TPM and the body temperaturewas maintained at 37�C–38�C with a heated pad (KentScientific).

An inverted confocal microscope (model IX81, OlympusAmerica) was modified to conduct TPM, as previouslydescribed (30). As a laser source, a tunable Ti:sapphirefemtosecond laser (Chameleon Ultra II, Coherent LaserGroup) was used, and the power was modulated using acombination of neutral density filters (Chroma Technolo-gy). A beam expander (LSM Technologies) was used tomodulate the size of the beam that was then directed intoa scanning head (Fluoview 1000, Olympus America). Theemitted signal was aimed into a custom-made set of 3nondescanned detectors (LSM Technologies). Two dichroicmirrors and the barrier filters were purchased fromChromaTechnology, and 3 cooled photo multipliers (PMT) werepurchased from Hamamatsu. The first PMT (510-nmdichroic mirror, 400- to 480-nm barrier filter) detected theendogenous fluorescence and the second harmonic signal.GFP, venus, and Alexa 488 signal were detected with thesecond PMT (570-nm dichroic mirror, 505- to 560-nmbarrier filter). TagRFP and Alexa 594 were detected on thethird PMT (590- to 650-nm barrier filter). For time-lapseimaging of the live animals, the acquisition speed was set to0.3 frames per second. All images andmovies were acquiredusing a UPLSAPO�60, NA 1.2, a XLUMPFL20XW�20, NA0.95, and a XLPlanN �25, NA 1.05 water immersionobjectives (Olympus America).

Image processingThe image background noise was reduced by applying a 2

� 2 pixel low-pass filter to each image for 1 or 2 roundsusing Metamorph software (Molecular Devices). Bright-ness, contrast, and gamma correction were applied. For themovies, the alignment of each frame was adjusted usingImageJ (W. Rasband, NIH) with Stackreg plug-in. Volumerenderingwas conducted using Imaris 7.4 64-bit (Bitplane).The final preparation of the images was managed withAdobe Photoshop CS. Movies were assembled with Meta-morph and compressed with Quicktime Pro.

Western blottingCells or small pieces of tissues were rinsed with PBS and

rapidly lysed with protein lysis buffer (62.5 mmol/L Tris-HCl (pH 6.8), 2% SDS, 10% glycerol, 50 mmol/L dithio-threitol), and transferred immediately to microcentrifugetubes and sonicated for 20 seconds. Protein yield was

quantified using the Quick start Bradford protein assay(Bio-Rad). Equivalent amounts protein (80–100 mg) wereseparated by SDS-PAGE, transferred to polyvinylidenedifluoride membranes, and immunodetection conductedwith tubulin as loading control.

In vitro wound scratch assayTumor cells were seeded uniformly (2 � 105 cells/well)

onto a silicone culture insert (Ibidi) and grown to 100%confluence as described by the manufacturer. After 24hours, the inserts were removed. Images of the woundswere acquired 0, 3, 6, 9, and 12 hours after the onset by amicroscope (Olympus), equipped with a charged-coupleddevice camera. The wound closure distance was measuredand reported.

Cell-cycle analysisCell-cycle analysis was conducted as described (24).

Briefly, cells were harvested in DMEM with 1% FBS, fixedin 70% ethanol, and stained with 50 mg/mL of propidiumiodide and 0.1 mg/mL of RNaseA in PBS. DNA content ofcells was quantified on a Becton Dickinson FACScan andanalyzed using Cell Quest software (Immunocytometrysystem; Becton Dickinson).

Tumor microvessel density analysisTumor microvessels were identified by CD31 and Lyve1

immunofluorescent staining of blood and lymphatic ves-sels, respectively. Images were captured from random areasin each section and morphometric analysis was conductedusing the MetaMorph 4.0 Imaging system (MolecularDevices Corporation). Microvessel density was calculatedby dividing the total perimeter of microvessels by thenumber area counted. Values reported for each experimen-tal condition correspond to the average values obtainedfrom 8 random field images in 2 independent experiments.

Immunofluorescence analysis for tumor cellproliferation and apoptosis

Tumor cell proliferation and apoptosis was estimated byusing anti-Ki67 and cleaved-caspase-3 antibody, respective-ly. Images were captured from at least 5 different areas ineach tissue slide. After adjusting the fluorescent signal–noise threshold of the images, the total area presentingfluorescent signalwasmeasuredusing ImageJ (NIH). Valuesreported correspond to the mean � SE of values obtainedfrom 4 samples for each experimental condition.

Analysis of F-actin levels at the plasma membrane invivo

The levels of F-actin at the plasma membrane wereevaluated in 2 experimental conditions. First, in cryosec-tions from xenografts of either Hela-O3-v or Hela-O3-vRab25 those were labeled with Alexa 594-phalloidin.Alternatively, in xenograft of Hela-O3 cells expressingRFP-lifeact and either venus or vRab25. Images of individ-ual cells were acquired using confocal microscope asdescribed above. The shape of the cells was determined by






the expression of venus or vRab25. A region of interest wasdelineated around the cell surface (approximately 1 mmfrom each side of the cell border) and the integratedfluorescence intensity was calculated by using Image J. Atleast 8–10 cells per experimental conditionswere evaluated.

Statistical analysisThe statistical analysis was conducted using Prism 5

(GraphPad Software). For each experiment, the statisticaltests were indicated in the result sections. Unpaired t-test (2-tailed) and one-way ANOVA were used to analyze the IHCstaining scores and the 3D invasion assays. Fisher’s exact testand x2 test were used for the tongue cancer metastasis invivo. Statistical significance was calculated at a 95% confi-dence level.

ResultsRab25 is downregulated in head and neck squamouscell carcinomaGenetic evidence supports the role of Rab25 in the

development and progression of various cancers in humanand mouse models including ovarian, breast, and prostatecancers (20, 21, 23, 31, 32). Recently, loss of Rab25 wasreported to increase the incidence of colorectal carcinomain mice (33). Therefore, we conducted IHC on humanHNSCC tissue arrays to determine whether Rab25 wasabnormally expressed. In healthy individuals, we foundthat the staining intensity of Rab25 correlated with thedegree of cell differentiation. Indeed, there is stronger

immunoreaction of Rab25 in the upper mucosal layers(more differentiated cells, Fig. 1, asterisk) than in the basaland suprabasal layers (less differentiated cells, Fig. 1, arrow-head). Moreover, we found that Rab25 expression wassignificantly lower in both advanced andmetastatic tumorswhen compared with tissues from healthy individuals (Fig.1A andB).However, Rab25 expressionwas not significantlydifferent among TNM stages. Moreover, we determined theexpressionofCLIC3, a protein linked toRab25 inpancreaticcancer (32). Surprisingly, we found high levels of CLIC3expression in tumors and normal tissue regardless of theRab25 expression (Supplementary Fig. S1A). These findingssuggest that Rab25 may operate in HNSCC through adifferentmechanism from those proposed for other tumors.Finally,we found thatRab25 expressionwasdownregulatedin cervical cancer in a similar fashion to HNSCC tissues(Supplementary Fig. S1B).

Rab25 downregulation promotes cell invasion in a 3Dmicroenvironment

To investigate the role of Rab25 downregulation inHNSCC, we screened a panel of human oral squamouscarcinoma cell lines for Rab25 expression (Fig. 2A). All thetested cell lines expressed Rab25 at levels comparable tonormal oral keratinocytes (Fig. 2A). Therefore, we engi-neered HN12 cells to stably expressed 3 different shRNAstargeting Rab25 (HN12-shRab25) or a scrambled shRNA(HN12-shScramb) as a control (Fig. 2B). This cell line waschosen because it forms tumors in orthotopic models of

Figure 1. Rab25 expression inhuman HNSCC. A, representativecore tissues from a human oralcancer tissue array stained using anantibody directed against humanRab25. Tissues were groupedaccording to the TNM classification(T1, T2, T3, T4, and Met). Rab25immunoreaction in moredifferentiated (asterisk) and lessdifferentiated (arrowhead) cells. Met,metastasis. The insets show highermagnification of the staining patternof cancer cells. B, quantitativeanalysis of Rab25 expression fromhuman normal and oral cancer tissuearray. ��, P < 0.0001; one-wayANOVA, n¼ number of tissue cores.

AT2Normal T4 Met

*

B10

n = 23

*** ***

8

sc

ore ns

4

6 n = 169

n = 10

T3; n = 21

T4; n = 29

Sta

inin

g

0

2

T1; n = 70

T2; n = 49

Normal Primary Met






A

B

Venus

v v-Rab25

Rab25

Tubulin

HeLa-O3

C

Inv

as

ion

(µ

m)

ns ns

***

Ha

Ca

T

C3

3A

SiH

a

He

La

-J

He

La

-A

He

La

O3

Ca

l27

UM

SC

C2

OR

L48

OR

L1

50

HN

12

UM

SC

C1

7B

Rab25

EGFR

Tubulin

HN

OK

(co

ntr

ol)

sh-Scramb sh-Rab25-1

Med

ium

+

EG

F +

Seru

m

xz view

v v-Rab25

Med

ium

+

EG

F +

Seru

m

xz view

D

Venus

Tubulin

HeLa-O3

v-R

ab

25

v-R

ab

25

-11

v-R

ab

11-2

5

v

Scramb Rab25

-1

ns

Inv

as

ion

(µ

m)

*** ***

v v-Rab25

0

20

40

60

80

100

Medium +Serum +EGF

Inv

as

ion

(µ

m)

ns

v v-

Rab25

v-Rab25-

11

v-Rab11-

25

ns

*** *** *** ***

ns

171 213

Rab25

Rab11a

Rab25-11

Rab11-25

N-termini amino acid

0

20

40

60

80

Medium +Serum +EGF

*** ***

*** *

Rab25

-2

Rab25

-3

shRNA

Rab25

Tubulin

HN12

Scra

mb

shRNA

Rab

25-1

Rab

25-2

Rab

25-3

0

40

80

120

160

Media +Serum +EGF

216 170

169

170 1

1

Figure 2. Rab25 controls tumor cell invasion in a 3D in vitro model. A, the protein levels of RAB25 and EGFR were determined by Western blot analysis in acancer cells panel using human normal oral keratinocytes (HNOK) as control. Tubulin was used as loading control. B and C, invasion assay in collagentype I matrix. HN12 cells were engineered to stably express shRNA for RAB25 (sh-Rab25) or control scramble shRNA (sh-Scramb; B, left, Westernblot analysis), whereas Hela-O3 cells were engineered to express venus-RAB25 (v-Rab25) or venus (v) as control (C, left, Western blot analysis). D, Westernblot analysis and collagen matrix invasion assay of Hela-O3 cells expressing v, v-Rab25, vRab25-11, and vRab11-25. EGF (10 ng/mL) or serumwas used aschemoattractants. The volume rendering from Z-stacks was conducted using Imaris (Bitplane). The extent of invasion was assessed by measuring thedistanceof the cell front from the topof thematrix thatwasmarkedbyfluorescent beads.Data representmean�SEM from8 random fields. ns, not statisticallysignificant; �, P < 0.01; ��, P < 0.0001; one-way ANOVA in B and D; unpaired t-test in C; scale bars, 20 mm.






oral cancer (28). Downregulation of Rab25 did not affecteither cell proliferation or motility when cultured on dif-ferent substrates in two-dimension (2D; data not shown).However, HN12-shRab25 invaded a 3D collagen matrix inthe presence of EGF, whereas HN12-shScramb and HN12parental line did not (Fig. 2B). To further characterize thisfinding, we screened additional human cancer cell lines forthe expression of Rab25 (Fig. 2A). Among the cell lines thatdid not express Rab25,we selectedHeLa-O3 cells, which arehighly metastatic when injected into floor of the mouth ofimmunocompromised mice [this is a variant of HeLa ade-nocarcinoma cells formerly mistaken for an oral cancer cellline (OSCC3; ref. 34)]. This cell line also expressed highEGFR level, which is commonly upregulated in HNSCC(Fig. 2A; ref. 35). HeLa-O3 cells were engineered to stablyexpress either Rab25 fused with the fluorescent proteinVenus (HeLa-O3-vRab25) or Venus alone (HeLa-O3-v;Fig. 2C). Notably, we observed that HeLa-O3-v cells migrat-ed through a 3D collagen matrix when stimulated by eitherserum or EGF without effecting cell proliferation, whereasHeLa-O3-vRab25 did not (Fig. 2C). Rab25 and Rab11ashare significant sequence homology with the exception ofthe c-terminal hypervariable region, which contains thetargeting information for the Rab proteins (6, 9, 10). Toconfirm that this effect is specific for Rab25, we constructed2 chimeras fusedwith venus: one, inwhich the c-terminusofRab25 was replaced with Rab11a c-terminus (vRab25-11),and the other, in which the c-terminus of Rab11a wasreplaced with Rab25 c-terminus (vRab11-25; Fig. 2D).Notably, cells expressing vRab11-25 (Hela-O3-vRab11-25) were impaired to migrate through the 3D collagenmatrix, whereas cells expressing vRab25-11 (Hela-O3 v-Rab25-11) were not (Fig. 2D). These results suggest thatalteringRab25 expression is sufficient to change the invasivebehavior of tumor cells in an in vitro 3Dmicroenvironment.

Rab25 reexpression inhibits metastasis to cervicallymph nodes in a mouse model for oral cancerTo assess whether Rab25 is involved in the development

and progression of SCC in vivo, we used a previously estab-lished tumor xenograft model in which tumor cells areinjected into the tongue of immunocompromised mice(36). HN12 cells formed primary tumors but did not invadelymphnodes evenafter 60days (Fig. 3A).On theotherhand,HeLa-O3cells formedavisible tumormasswithin7–10days(Fig. 3B, top) and invaded the cervical lymph nodes inmorethan 80% of the mice within 3 weeks (Fig. 3B, bottom).When these cells were injected into the flank of immuno-compromisedmice, tumor formation occurred but they didnot invade or metastasize (data not shown). Interestingly,other variants of HeLa cells did not metastasize whentransplanted into the tongue (Fig. 3B) or in other areas,such as the flank or the cervix (data not shown).Next, we transplanted HeLa-O3-vRab25 or HeLa-O3-v

cells into the mouse tongue to investigate whether Rab25has a role in the observed cervical lymph nodes metastasis.The expression of Venus or Venus-Rab25 did not affect theability of HeLa-O3 cells to form primary tumors in the

tongues of immunocompromised mice (Fig. 3C). Striking-ly, themetastasis incidence inHeLa-O3-vRab25 tumorswassignificantly reduced compared with the incidence inHeLa-O3-v tumors (Fig. 3C). This reduction was not because ofdifferences in tumor growth (assessed by tumor weight andvolume), cell proliferation (assessed by Ki67 staining), orapoptosis (assessed by cleaved-caspase-3 staining; Supple-mentary Fig. S2A and S2C, respectively). Moreover, theexpression of Rab25 did not affect the formation of bloodand lymphatic vessels, in contrast to a previous report inbreast cancer (Supplementary Fig. S2B; ref. 19). The effectsfrom Rab25 were specific, as shown by the transplantationof Hela-O3-vRab11-25 and Hela-O3-vRab25-11. Indeed,the expression of vRab11-25 reduced the metastatic inci-dence in the tongue xenograft tumor, whereas the expres-sion of vRab25-11 recuperated the cervical metastasis inci-dence in nearly 50% of the mice (Fig. 3D).

Unfortunately, we could not use HN12-shRab25 cells invivo, because Rab25 protein expression returned back to itslevel after injection into the tongue in spite of stableexpression of the shRNA (data not shown).

Next, we determined whether Rab25 would affect inva-sion and metastasis in vivo, similar to what we observed inan in vitro 3D assay. To this aim, we used intravital micros-copy, which enables imaging the dynamics of biologicprocesses in live animals (37). In addition to the other celllines, we generated HN12 and HeLa-O3 cells stably expres-sing histone 2B fused with GFP (H2B-GFP) for a bettervisualization of the tumor mass. Moreover, we developed atongue-holder device to image tumors daily and to mini-mize themotion artifacts from theheartbeat and respiration(Fig. 4A). This approach has 2main advantages: first, it doesnot require invasive surgical procedures, thus making pos-sible to conduct longitudinal studies, and second, it facil-itates repeated imaging in the same area. We acquiredimages of primary tumors located 100–200 mm below thetongue surface by using a combination of TPM and secondharmonic generation (Fig. 4B; Supplementary Movies S1and S2). The tumors edges were imaged once a day for 15–25 consecutive days. Both the depth and the orientation ofthe tumors were determined by reference to collagen fibers(Fig. 4B), which have a very slow turnover and their patternis maintained over the monitoring period (38; Fig. 5A,arrows). Interestingly, we observed the tumors expandingin size and individual cells departing from the tumor mass(Fig. 5A, arrowheads). To catch themovement of these cells,the tumors edges were imaged by consecutive Z-scans at 5minutes intervals (Fig. 5B). Strikingly, we observed somecellsmoving away from the edges of the tumors at a speed ofapproximately 1–2 mm per minutes (Fig. 5B, arrowheads,and Supplementary Movie S3), whereas the majority of thecells did not migrate within the observation period. Later,these tumor cells were either found near or inside lymphaticvessels, as revealed by intraorgan injections of fluorescentdextran (Fig. 5C and D; ref. 39). Later, tumor cells wereeither completely encapsulated once they reached andcolonized the cervical lymph nodes (Fig. 5E) or they scat-tered near large blood vessels (Fig. 5F).






% o

f m

ice

wit

h

tum

or

bu

rde

n

A

C

B

% o

f m

ice

wit

h

tum

or

bu

rde

n

% o

f m

ice w

ith

tum

or

bu

rden

0

20

40

60

80

100

Primary LN Met

0

20

40

60

80

100

Primary LN Met

0

20

40

60

80

100

Primary LN MetTongue

Dorsal

Tongue

Tongue

Lymph node

HN12 (n = 12)

HeLa-O3 (n = 8)

HeLa-A (n = 8)

HeLa-O3-v (n = 11)

HeLa-O3-vRab25 (n = 12)

HN12

HeLa-O3

HeLa-O3-v

0

20

40

60

80

100

Primary LN Met

v-Rab25 (n = 8)

v-Rab25-11 (n = 10)

v-Rab11-25 (n = 10)

v- (n = 8)

% o

f m

ice w

ith

tum

or

bu

rden

D HeLa-O3

***

***

ns

*** ns

***

ns

ns ns

***

**

Figure 3. Rab25 controls tongue cancer metastases in vivo—tumor cell lines (HN12 in A; HeLa-O3 and HeLa-A in B; HeLa-O3-v and HeLa-O3-vRab25 in C, and Hela-O3-v, -v-Rab25, -v-Rab25-11, and -v-Rab11-25 in D) were transplanted into the lateral tongues of immunocompromisedmice. Mice were euthanized after 60 days. Tongue primary tumors and cervical lymph nodes were removed for histopathologic evaluation.Histology of whole tongue tumors and metastasis in cervical lymph node are depicted from representative tissue for each tumor cell (A, HN12; B,HeLa-O3; and C, HeLa-O3-v). Broken lines indicate the tumor mass border. High magnification insets are in the right panels. Quantitative datarepresent percentage of mice with tumor burden or cervical lymph nodes metastasis. LN Met, lymph nodes metastasis; n ¼ number of mice.Quantitative analysis of tumor burden and LN Met are shown in graphs. ns, not statistically significant; ��, P < 0.001; ��, P < 0.0001; Fisher exact testin A, B, and C; x2 test in D.






Todeterminewhether expression of Rab25directly affect-ed the motility of the tumor cells in vivo, we co-injectedHeLa-O3 cells stably expressing the RFP (HeLa-O3-RFP)withHeLa-O3-vRab25 and imaged the tumor edge either ona daily basis or at shorter intervals (every 5 minutes). Bothcell lines were equally distributed in the tumor mass in theearly stage (Fig. 6A). However, HeLa-O3-RFP cells migratedfrom the primary sites, which became enriched inHeLa-O3-vRab25 cells at later time points (Fig. 6B). Moreover, wecould not observe any HeLa-O3-vRab25 cells leave theprimary site (data not shown). Consistent with this finding,HeLa-O3-RFP cells were the only cells that invaded thelymphatic system and metastasized to the cervical lymphnodes (Fig. 6C).

Rab25 reexpression blocks the formation of actin-richprotrusive structuresAs shown above, reexpression of Rab25 negatively reg-

ulates the ability of tumor cells tometastasize to the cervical

lymph nodes in our mouse model of oral cancer. Our dataalso suggest that expressionof Rab25 inhibits themotility ofa subpopulation of tumor cells. To gain some insight on themachinery regulated by Rab25 we imaged both HeLa-O3-vand HeLa-O3-vRab25 tumors at higher resolution andfaster acquisition speed. Rab25 was clearly localized insmall and dynamic vesicles scattered in the cytoplasm,confirming that its expressiondidnot elicit any toxic cellulareffect (Fig. 7A, Supplementary Movie S4). Interestingly,HeLa-O3-v showed a series of dynamic cell protrusionsthat were not detected in HeLa-O3-vRab25 (Fig. 7A, Sup-plementary Movie S4). Because the actin cytoskeleton hasbeen implicated in the formation of various plasma mem-brane structures involved in cell migration (5), we investi-gated whether F-actin associated with these protrusions.Indeed, actin was enriched at the plasma membrane, par-ticularly in the protrusions, in both 3D collagenmatrix andlive animals (Fig. 7B and C, Supplementary Movie S5).Notably, cells expressing Rab25 had a substantial reduction

A

B

8 µm 56 µm 106 µm 132 µm

8 µm 56 µm 106 µm 132 µm xy

Exc. 750 nm

Exc. 930 nm

yz

8 µm

56 µm

106 µm

132 µm

Tumor cells

(H2B-GFP)

Muscle fibers

Tongue

epithelium

8 µm

56 µm

106 µm

132 µm

Top view

Side view Tongue

Tumor mass

Tongue device

Figure 4. Intravital imaging of the tongue cancer model. A, schematic drawings of the tongue holding device, animal set up for intravital imaging, and primarytumor mass growing in the tongue. B, intravital microscopy of tongue cancer at the primary site. Hela-O3 cells expressing H2B-GFP were transplantedinto the tongue submucosa. Before imaging, a 70 kDa dextran was injected systemically to reveal stromal cells. A Z-scan was conducted by using TPM(60� water immersion lens, NA 1.2, Olympus). The same area was imaged using either 750 nm (top) or 930 nm (bottom) as excitation wavelengths. Bothconditions revealed tumor cells (green) and stromal cells (red). However, the excitation at 750 nm revealed the tongue parenchyma (cyan), whereas theexcitation at 930 nm revealed components of the extracellular matrix and myosin filaments from the muscle fibers (cyan). Scale bars, 20 mm.






in the levels of F-actin at the plasma membrane (Fig. 7B–E,SupplementaryMovie S5). Moreover, we found that knock-ing down Rab25 in HN12 cells also resulted in the actin-enriched protrusion formation in 3D collagen matrix (datanot shown). To address whether the actin cytoskeletonaffected tumor cells invasion, we used 2 potent F-actindisrupting drugs (cytochalasin D and latrunculin A) in the3D invasion assays. Both inhibitors significantly affectedtumor cells invasion independently of the Rab25 levels (Fig.7F). Interestingly, modulations of Rab25 levels in bothHeLa-O3 and HN12 cells did not change the arrangementof the cytoskeleton when cells were grown in 2D substratessuggesting that the 3D architecture and the tissue microen-vironment play a fundamental role in this process (data notshown).

DiscussionSeveral evidences suggest that deregulation of molecules

involved in endocytosis plays a fundamental role in humandiseases including cancer. Indeed, endocytosis coordinatesmultiple steps that may affect cancer progression, such asreceptor-mediated signaling from the plasma membrane,cell-to-cell contact, and cell motility (7). Rab proteins areconsidered master regulators of endocytosis and oftenderegulated in various cancers. Among the Rab familymembers, Rab25 has been implicated in different cancersalthough its mechanism of action has been controversial.Indeed, both upregulation and downregulation of Rab25have been associated with tumor growth, differentiation,and poor prognosis. Moreover, Rab25 has been involved inseveral cellular processes such as proliferation, apoptosis,

B

C D

A

E

yz

xy

xyD7 D8 D9

0

5

10

15

20

25

F

Figure 5. Long-term imaging of the tongue cancer. A–F, Hela-O3 cells expressing either venus (A and B) or H2B-GFP (C–F) were transplanted into the tonguesubmucosa, as described in Materials and Methods. A, after 7 days from the injection, the edges of the tumor were imaged on a daily basis by using TPM(excitation wavelength of 930 nm) to reveal collagen fibers (red) and the tumormass (green). Z-stacks were acquired at day 7 (D7), 8 (D8), and 9 (D9) by using a25�water lens (NA 1.05, Olympus). 3D reconstructions (top, yz view) andmaximal projections of the xy viewwere conducted by using Imaris (Bitplane;middleand bottom). The extracellular matrix at the surface of the tongue was used as a reference point (middle, arrows). Individual tumor cells were observedmigrating from the tumormass (arrowheads). Bar, 50 mm.B, after 14 days, the edge of the tumorwas imaged as described above by conducting Z-stacks for 2hours at 5 minutes interval. Individual cells were observed leaving the tumor mass (insets, red arrowheads). Bar, 20 mm. C and D, Texas Red-dextran (70 kDa)was injected in the tongue tomap the lymphatic vessels, and Z-scanwere conducted, as described above. Maximal projections show tumor cells expressingH2B-GFP (green) either in proximity (C) or inside lymphatic vessels (D). Note collagen fibers in C (cyan). E and F, dextran (70 kDa) was intravenously injected tolabel blood vessels and stromal cells, and cervical lymph nodes were exposed and imaged as described above. E, tumor cells (green) colonized a cervicallymph node: stromal cells (red) and collagen fibers (cyan). F, tumor cells (green) are in close proximity to a blood vessel (red). Bars, 50 mm.






angiogenesis, cell cycle, trafficking of adhesion molecules,and cell migration (19–21, 31, 40–43). In this study, wedetermine the role of Rab25 in HNSCC formation andmetastasis. To this aim, we used a series of approaches,which include expression analysis of HNSCC patient tis-sues, analysis of cellular function in 3Dmicroenvironmentsin vitro, and tumor cells dynamics in vivo using intravitalmicroscopy in a tongue cancer model.First, we found a significant reduction in Rab25 expres-

sion in HNSCC patient tissues, although without any cor-relation between Rab25 expression and tumor stage. Thesefindings would favor a model in which Rab25 acts as atumor suppressor, as suggested for other cancers (19, 22).However, loss of Rab25 alone may not be sufficient toinitiate tumorigenesis because Rab25 knockout mice donot develop any spontaneous tumors (33). In addition, wedetected high levels of CLIC3 expression in both normaland HNSCC tissues independent of Rab25 expression. Thisobservation is in contrast to a report suggesting that CLIC3and Rab25 synergistically drive tumor progression in pan-creatic ductal adenocarcinoma (32). Our data stronglysuggests that Rab25 may operate in HNSCC differentlyfrom what was described for other cancers (19–21, 23, 31, 32, 40, 44, 45). Indeed, we found that Rab25downregulation promotes cell invasion and metastasis inHNSCC. The main evidence comes from an invasion assayin a 3D collagen matrix and in a tongue cancer mousemodel. We used 2 cancer cell lines; HN12 (Rab25 levelscomparable to normal oral keratinocytes) and Hela-O3(expression of Rab25 below the detection levels) that wereengineered to knockdown Rab25 or express fluorescently

tagged Rab25, respectively. Both cell lines were chosen asthey form aggressive tumors when implanted in the tongueor the floor of mouth in immunodeficient mouse (34, 36).We found that alteration in Rab25 expression did not affectcell proliferation or apoptosis both in vitro and in vivo.However, cells lackingRab25migrated through3Dcollagenmatrix when stimulated by EGF or serum. Consistent withthis finding, HeLa-O3 cells metastasized to cervical lymphnodes when transplanted in the tongue or in floor ofmouth(34), whereas other HeLa variants, which lacked bothRab25 and EGFR, did not. This observation suggests apossible link between Rab25 and EGFR signaling in cellmigration and metastasis. Moreover, we ruled out the rolesof Rab25, in tumor growth or vascular angiogenesis regu-lation, as reported in breast cancer (19, 42).

To gain further insight on how Rab25 regulate invasionand metastasis in vivo, we used intravital microscopy. Weobserved that a subpopulation of cells lacking Rab25migrated from the tumor mass, whereas Rab25-expressingcells were confined to the primary site. Consistent with thisobservation, only Rab25-depleted cells reached lymphaticvessels, intravasated, and spread to the distal lymph nodes.These observations suggest that tumor invasion occursthrough single cell rather than collective migration in thismodel (46) and this process is facilitated by the lack ofRab25. Finally, we observed that reexpression of Rab25induced significant changes in cell shape.During cellmigra-tion, the actin cytoskeleton is rearranged at the cell surfaceand forms a series of structures such as invadopodia, podo-somes, lamellipodia, and ruffles that interact with theextracellular matrix components (46). Both in vivo and in

A B

C

D17 D18 D19D9

Figure 6. RAB25expression inhibits lymphaticmetastasis in tongue cancer—HeLa-O3cells expressingRFP (red) or venus-Rab25 (green)were cotransplantedinto the lateral tongue of immunocompromised mice. A and B, the edges of the tumor were imaged on a daily basis by intravital TPM, as described inFig. 5. A, tumor mass after 9 days (D9). B, maximal projections of the tumor mass imaged at day 17 (D17), 18 (D18), and 19 (D19). Note that the tumor cellslacking Rab25 (red) migrate from the tumor mass that is mainly formed by cells expressing Rab25 (green). Bars, 50 mm. C, after 30 days, tumors andcervical lymphnodeswere removed, fixed, andprocessed asdescribed inMaterials andMethods. Tissue cryosectionswere labeledwith an antibody directedagainst Lyve-1, a marker for lymphatic vessels (cyan). Primary tumors are shown in the left and middle panels. Note that only cells lacking Rab25 are able toinvade the lymphatic vessels (middle) and colonize the cervical lymph node (right).






a 3D matrix, cells lacking Rab25 displayed actin-rich pro-trusions, which has similar morphology to invadopodia(47, 48). Rab25 expression significantly reduced the num-ber of actin-rich structures and the overall levels of F-actin at

the plasma membrane, suggesting a possible involvementin actin dynamics regulation. Moreover, we found that theinvasion of tumor cells in a 3D collagen assay is significantlyinhibited by the pharmacologic disruption of the actin

Ph

allo

idin

P

ha

llo

idin

L

ife

ac

tGF

P

C

B

D

A

HeLa-O3-RAB25 HeLa-O3

0

20

40

60

80

100

120

v v-Rab25

Medium +EGF +EGF-latA +EGF-cytoD

Invasio

n (

µm

)

*** ***

*** ***

Phalloidin LifeactActi

n p

ixel in

ten

sit

y p

er

cell

(arb

itra

ry u

nit

)

* ***

E

F

HeLa-O3 HeLa-O3--Rab25

0

1,000

2,000

3,000

Figure 7. RAB25 expression reduces actin-rich protrusion in tumor cells. A, intravital TPM of tongue tumors from HeLa-O3 expressing venus (left)or venus-Rab25 (right). Cells lacking Rab25 exhibit dynamics protrusions (arrowheads and Supplementary Movie S4). B, HeLa-O3 cells expressingvenus (left) or venus-Rab25 (right) were allowed to invade a 3D collagenmatrix, fixed, and labeledwith phalloidin. In cells lackingRab25, actin-rich protrusionsare clearly visible at the cell surface. C, mouse tongue xenograft tissues from HeLa-O3 expressing venus (left) or venus-Rab25 (right) were stainedwith phalloidin (overlay) and phalloidin alone (bottom). Broken lines indicated the approximate tumor cell borders. Arrowheads indicate the thick actinstructures at cell protrusions. D, intravital imaging of tumor cells expressing lifeactGFP inHeLa-O3expressingRFP (left) or RFPRab25 (right) in tongue tumors.Arrowheads indicate the thick actin structures at cell protrusions. Scale bars, 20 mm. E, quantification of F-actin at the plasma membrane. Actin expression(phalloidin staining and Lifeact expression) from HeLa-O3 and HeLa-O3-Rab25 cells was analyzed by ImageJ software as described in Materials andMethods. The graphs represent the pixel intensity at the plasma membrane of each cell (mean� SE). �, P < 0.01; ��, P < 0.0001; unpaired t-test. F, collageninvasion assay of HeLa-O3-v cells stimulated with EGF (10 ng/mL) in the presented or absent of actin polymerization inhibitors (10 mmol/L of cytochalasinD-cytoD and 1 mmol/L of latrunculin A-latA). Cell invasion quantification was conducted as described in Fig. 2. Data represent mean � SEM from 8 randomfields. ns, not statistically significant; �, P < 0.01; ��, P < 0.0001; one-way ANOVA.






cytoskeleton. Thus, we can hypothesize that Rab25 mayregulate trafficking of molecules that negatively regulateactin assembly (e.g., signaling molecules, integrin, or cho-lesterol-rich membranes; refs. 5, 47, 48), although wecannot rule out other mechanisms. This idea is supportedby the role of FIP3, a protein that simultaneously interactswith the small GTPase Arf6 and Rab25/Rab11, which con-trols the motility of breast cancer cells by regulating Rac1and actin dynamics (49). On the basis of the fact that Rab25is localized in intracellular vesicles and that Rab25 has beenimplicated in endosomal recycling (50), we speculated thatthe lack of Rab25might prevent these actin regulators fromreaching the plasma membrane and divert them to degra-dation pathway. Indeed, a similar mechanism has beensuggested in breast and pancreatic cancer (6, 32). However,in these models, overexpression of Rab25 enhanced cellmigration and invasion by increasing the recycling of integ-rin b1 from a late endosomal compartment to the plasmamembrane (6, 32). These differences in the modality ofaction of Rab25 may be explained by the fact that thetrafficking of molecules involved in migration may occurthrough different pathways in which Rab25 may interactwith a completely different set of molecules.In conclusion,wehave shown that the downregulation of

Rab25 plays a significant role in HNSCC and suggested thatRab25may regulate tumor invasion andmetastasis throughactin remodeling at the plasma membrane. We believe thatidentification of the molecules interacting with Rab25 andelucidation of the molecular machinery underlying its

action may provide new strategies to prevent metastasis inoral cancer patients.

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

Authors' ContributionsConception and design: P. Amornphimoltham, J.S. Gutkind, R. WeigertDevelopment ofmethodology: P. Amornphimoltham, A.Masedunskas, K.Leelahavanichkul, V. Patel, R. WeigertAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): P. Amornphimoltham, K. Rechache, J. Thomp-son, A. Masedunskas, K. Leelahavanichkul, A. Molinolo, R. WeigertAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): P. Amornphimoltham, K. Rechache, A.Masedunskas, R. WeigertWriting, review, and/or revision of the manuscript: P. Amornphimol-tham, K. Rechache, A. Molinolo, J.S. Gutkind, R. WeigertAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): P. Amornphimoltham, J.S. Gut-kind, R. WeigertStudy supervision: P. Amornphimoltham, R. Weigert

AcknowledgmentsThe authors would like to thank Dr. Porat-Shliom and Mr. Mohibullah

Tora for critical reading of the manuscript.

Grant SupportThis research was supported by the Intramural Research Program of

National Institute of Dental and Craniofacial Research, NIH.The costs of publication of this article were defrayed in part by the

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

Received September 4, 2012; revisedDecember 20, 2012; accepted January7, 2013; published OnlineFirst January 22, 2013.

References1. Gale N PB, Sidransky D, Westra W, Califano J. Tumours of the

hypopharynx, larynx and trachea (Epithelial precursor lesions). In:Barnes L EJ, Reichart P, Sidransky D, editors. World Health Organi-zation classification of tumours pathology & genetics head and necktumours International Agency for Research on Cancer (IARC). Lyon:IARC Press; 2005. p. 140–3.

2. Haddad RI, Shin DM. Recent advances in head and neck cancer.N Engl J Med 2008;359:1143–54.

3. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA CancerJ Clin 2012;62:10–29.

4. Valastyan S, Weinberg RA. Tumor metastasis: molecular insights andevolving paradigms. Cell 2011;147:275–92.

5. Ridley AJ. Life at the leading edge. Cell 2011;145:1012–22.6. Caswell P, Norman J. Endocytic transport of integrins during cell

migration and invasion. Trends Cell Biol 2008;18:257–63.7. Mosesson Y, Mills GB, Yarden Y. Derailed endocytosis: an emerging

feature of cancer. Nat Rev Cancer 2008;8:835–50.8. Mitra S, Cheng KW,Mills GB. RabGTPases implicated in inherited and

acquired disorders. Semin Cell Dev Biol 2011;22:57–68.9. Colicelli J. Human RAS superfamily proteins and related GTPases. Sci

STKE 2004;2004:re13.10. Pereira-Leal JB, Seabra MC. Evolution of the rab family of small GTP-

binding proteins. J Mol Biol 2001;313:889–901.11. Stenmark H. Rab GTPases as coordinators of vesicle traffic. Nat Rev

Mol Cell Biol 2009;10:513–25.12. Shimada K, Uzawa K, Kato M, Endo Y, Shiiba M, Bukawa H, et al.

Aberrant expression of RAB1A in human tongue cancer. Br J Cancer2005;92:1915–21.

13. Chia WJ, Tang BL. Emerging roles for Rab family GTPases in humancancer. Biochim Biophys Acta 2009;1795:110–6.

14. Hou Q, Wu YH, Grabsch H, Zhu Y, Leong SH, Ganesan K, et al.Integrative genomics identifies RAB23 as an invasionmediator gene indiffuse-type gastric cancer. Cancer Res 2008;68:4623–30.

15. Liu YJ,WangQ, LiW, Huang XH, ZhenMC,HuangSH, et al. Rab23 is apotential biological target for treating hepatocellular carcinoma.WorldJ Gastroenterol 2007;13:1010–7.

16. Fukui K, Tamura S, Wada A, Kamada Y, Igura T, Kiso S, et al. Expres-sion of Rab5a in hepatocellular carcinoma: possible involvement inepidermal growth factor signaling. Hepatol Res 2007;37:957–65.

17. Croizet-Berger K, Daumerie C, Couvreur M, Courtoy PJ, van den HoveM-F. The endocytic catalysts, Rab5a and Rab7, are tandem regulatorsof thyroid hormone production. Proc Natl Acad Sci 2002;99:8277–82.

18. Calhoun BC, Goldenring JR. Rab proteins in gastric parietal cells:evidence for the membrane recycling hypothesis. Yale J Biol Med1996;69:1–8.

19. Cheng JM, Volk L, Janaki DK, Vyakaranam S, Ran S, Rao KA. Tumorsuppressor function of Rab25 in triple-negative breast cancer. IntJ Cancer 2010;126:2799–812.

20. Caswell PT, Spence HJ, Parsons M, White DP, Clark K, Cheng KW,et al. Rab25 associates with alpha5beta1 integrin to promote invasivemigration in 3D microenvironments. Dev Cell 2007;13:496–510.

21. Cheng KW, Lu Y, Mills GB. Assay of Rab25 function in ovarian andbreast cancers. Methods Enzymol 2005;403:202–15.

22. Goldenring JR, Nam KT. Rab25 as a tumour suppressor in coloncarcinogenesis. Br J Cancer 2011;104:33–6.

23. Cheng KW, Agarwal R,Mitra S, Lee JS, CareyM,Gray JW, et al. Rab25increases cellular ATP and glycogen stores protecting cancer cellsfrom bioenergetic stress. EMBO Mol Med 2012;4:125–41.

24. AmornphimolthamP, Patel V, Sodhi A, Nikitakis NG, Sauk JJ, SausvilleEA, et al. Mammalian target of rapamycin, a molecular target in






squamous cell carcinomas of the head and neck. Cancer Res2005;65:9953–61.

25. Hamid S, Lim KP, Zain RB, Ismail SM, Lau SH, Mustafa WM, et al.Establishment and characterization of Asian oral cancer cell lines as invitro models to study a disease prevalent in Asia. Int J Mol Med2007;19:453–60.

26. Patel V, Iglesias-Bartolome R, Siegele B, Marsh CA, LeelahavanichkulK, Molinolo AA, et al. Cellular systems for studying human oralsquamous cell carcinomas human cell transformation. In: Rhim JS,Kremer R, editors. Advances in experimental medicine and biology:human cell transformation. New York: Springer; 2012. p. 27–38.

27. Kanda T, Sullivan KF, Wahl GM. Histone €A��GFP fusion protein enablessensitive analysis of chromosome dynamics in livingmammalian cells.Curr Biol 1998;8:377–85.

28. AmornphimolthamP, Leelahavanichkul K,Molinolo A, Patel V, GutkindJS. Inhibition of mammalian target of rapamycin by rapamycin causesthe regression of carcinogen-induced skin tumor lesions. Clin CancerRes 2008;14:8094–101.

29. Charafe-Jauffret E, Tarpin C, Bardou VJ, Bertucci F, Ginestier C, BraudAC, et al. Immunophenotypic analysis of inflammatory breast cancers:identification of an 'inflammatory signature'. J Pathol 2004;202:265–73.

30. Masedunskas A, Weigert R. Intravital two-photon microscopy forstudying the uptake and trafficking of fluorescently conjugated mole-cules in live rodents. Traffic 2008;9:1801–10.

31. ChengKW, Lahad JP, KuoWL, LapukA, YamadaK, AuerspergN, et al.The RAB25 small GTPase determines aggressiveness of ovarian andbreast cancers. Nat Med 2004;10:1251–6.

32. Dozynkiewicz MA, Jamieson NB, Macpherson I, Grindlay J, van denBerghe PV, von ThunA, et al. Rab25 andCLIC3 collaborate to promoteintegrin recycling from late endosomes/lysosomes and drive cancerprogression. Dev Cell 2012;22:131–45.

33. NamKT, LeeHJ, Smith JJ, Lapierre LA, Kamath VP, Chen X, et al. LossofRab25promotes thedevelopment of intestinal neoplasia inmice andis associated with human colorectal adenocarcinomas. J Clin Invest2010;120:840–9.

34. HensonB, Li F, CoatneyDD,Carey TE,Mitra RS,KirkwoodKL, et al. Anorthotopic floor-of-mouthmodel for locoregional growth and spreadofhuman squamous cell carcinoma. J Oral Pathol Med 2007;36:363–70.

35. Molinolo AA, Amornphimoltham P, Squarize CH, Castilho RM, Patel V,Gutkind JS. Dysregulated molecular networks in head and neckcarcinogenesis. Oral Oncol 2009;45:324–34.

36. Patel V, Marsh CA, Dorsam RT, Mikelis CM, Masedunskas A, Amorn-phimoltham P, et al. Decreased lymphangiogenesis and lymph nodemetastasis by mTOR inhibition in head and neck cancer. Cancer Res2011;71:7103–12.

37. Weigert R, Sramkova M, Parente L, Amornphimoltham P, Maseduns-kas A. Intravital microscopy: a novel tool to study cell biology in livinganimals. Histochem Cell Biol 2010;133:481–91.

38. Brown E, McKee T, diTomaso E, Pluen A, Seed B, Boucher Y, et al.Dynamic imaging of collagen and itsmodulation in tumors in vivo usingsecond-harmonic generation. Nat Med 2003;9:796–800.

39. Melancon MP, Wang Y, Wen X, Bankson JA, Stephens LC, Jasser S,et al. Development of a macromolecular dual-modality MR-opticalimaging for sentinel lymph node mapping. Invest Radiol 2007;42:569–78.

40. Fan Y, Xin XY, Chen BL, Ma X. Knockdown of RAB25 expression byRNAi inhibits growth of human epithelial ovarian cancer cells in vitroand in vivo. Pathology 2006;38:561–7.

41. Bigelow RL, Williams BJ, Carroll JL, Daves LK, Cardelli JA. TIMP-1overexpression promotes tumorigenesis of MDA-MB-231 breast can-cer cells and alters expression of a subset of cancer promoting genesin vivo distinct from those observed in vitro. Breast Cancer Res Treat2009;117:31–44.

42. Yin Y, Shen F, Pei H, Ding Y, Zhao H, Zhao M, et al. Increasedexpression of Rab25 in breast cancer correlateswith lymphaticmetas-tasis. Tumor Biol 2012;33:1581–7.

43. Senga K, Mostov KE, Mitaka T, Miyajima A, Tanimizu N. Grainy-head-like 2 regulates epithelial morphogenesis by establishingfunctional tight junctions through the organization of a molecularnetwork among claudin3, claudin4, and Rab25. Mol Biol Cell 2012;23:2845–55.

44. Cheng JM, Ding M, Aribi A, Shah P, Rao K. Loss of RAB25 expressionin breast cancer. Int J Cancer 2006;118:2957–64.

45. Cheng KW, Lahad JP, Gray JW, Mills GB. Emerging role of RABGTPases in cancer and human disease. Cancer Res 2005;65:2516–9.

46. Friedl P, Alexander S. Cancer invasion and the microenvironment:plasticity and reciprocity. Cell 2011;147:992–1009.

47. Murphy DA, Courtneidge SA. The 'ins' and 'outs' of podosomes andinvadopodia: characteristics, formation and function. Nat RevMol CellBiol 2011;12:413–26.

48. Destaing O, Block MR, Planus E, Albiges-Rizo C. Invadosomeregulation by adhesion signaling. Curr Opin Cell Biol 2011;23:597–606.

49. Jing J, Tarbutton E, Wilson G, Prekeris R. Rab11-FIP3 is a Rab11-binding protein that regulates breast cancer cell motility bymodulatingthe actin cytoskeleton. Eur J Cell Biol 2009;88:325–41.

50. Casanova JE, Wang X, Kumar R, Bhartur SG, Navarre J, Woodrum JE,et al. Association of Rab25 and Rab11a with the apical recyclingsystem of polarized Madin-Darby canine kidney cells. Mol Biol Cell1999;10:47–61.






2013;19:1375-1388. Published OnlineFirst January 22, 2013.Clin Cancer Res Panomwat Amornphimoltham, Kamil Rechache, Jamie Thompson, et al. Rab25 Regulates Invasion and Metastasis in Head and Neck Cancer

Updated version

10.1158/1078-0432.CCR-12-2858doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2013/01/22/1078-0432.CCR-12-2858.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/19/6/1375.full#ref-list-1

This article cites 47 articles, 9 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/19/6/1375.full#related-urls

This article has been cited by 6 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/19/6/1375To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-12-2858

http://clincancerres.aacrjournals.org/content/suppl/2013/01/22/1078-0432.CCR-12-2858.DC1

http://clincancerres.aacrjournals.org/content/19/6/1375.full#ref-list-1

http://clincancerres.aacrjournals.org/content/19/6/1375.full#related-urls

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/19/6/1375


Documents

Rab25 Regulates Invasion and Metastasis in Head and Neck ...Advancing our understanding of the molecular mechan- ... may lead to more effective treatment options. Invasion and metastasis