Oral Oncology 50 (2014) 983–990

Contents lists available at ScienceDirect

Oral Oncology

journal homepage: www.elsevier .com/locate /ora loncology

Beclin1 inhibits proliferation, migration and invasion in tonguesquamous cell carcinoma cell lines

http://dx.doi.org/10.1016/j.oraloncology.2014.06.0201368-8375/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: Department of Oral and Maxillofacial Surgery,Guanghua School of Stomatology, Hospital of Stomatology, Sun Yat-sen University,Guangzhou, Guangdong 510055, China. Tel.: +86 2083862531; fax: +862083822807.

E-mail address: houjsgz01@yahoo.com (J. Hou).1 These authors contributed equally to this study.

Junquan Weng 1, Cheng Wang 1, Yawen Wang, Haikuo Tang, Jianfeng Liang, Xiqiang Liu,Hongzhang Huang, Jinsong Hou ⇑Department of Oral and Maxillofacial Surgery, Guanghua School of Stomatology, Hospital of Stomatology, Sun Yat-sen University, Guangzhou, Guangdong 510055, ChinaGuangdong Provincial Key Laboratory of Stomatology, Sun Yat-sen University, Guangzhou, Guangdong 510055, China

a r t i c l e i n f o

Article history:Received 26 January 2014Received in revised form 21 June 2014Accepted 30 June 2014Available online 2 August 2014

Keywords:Beclin1AutophagyProliferationInvasionMigrationTongue squamous cell carcinoma

s u m m a r y

Objectives: The role of autophagy is still a controversy in cancer development. In our previous study, weconfirmed that decrease of autophagy activity promotes malignant progression of tongue squamous cellcarcinoma (TSCC). However, the role of autophagy-related protein, Beclin1, has not well been docu-mented in TSCC. In this study, we aim to elucidate the role of beclin1 in TSCC progression and investigateits potential mechanisms.Materials and methods: TSCC cell lines, SCC9 and SCC15 were used to generate the stable cells with trans-fection lentivirus BECN1 and sh-BECN1. Then, Beclin1 expression was detected with qPCR and westernblot. Moreover, the expressions of autophagy-related proteins and tumor metastasis associated proteinswere examined by western blot and ELISA. For functional analysis, MTT assay were performed to evaluatethe proliferation activity and transwell assay was used to assess the migration and invasion ability.Finally, TSCC xenograft models were established to confirm the effect of Beclin1 on TSCC in vivo.Results: The results showed that BECN1 and sh-BECN1 virus transfection significantly increased ordecreased the mRNA and protein expression of Beclin1 in the transfected TSCC cells. Meanwhile, we alsoobserved that Beclin1 could enhance the expression levels of LC3-II, ATG4 and ATG5. Then, we revealedthat overexpression of Beclin1 inhibited proliferation, migration and invasion while knockdown ofBeclin1 promoted proliferation, migration and invasion in TSCC cells. Furthermore, we demonstrated thatvascular endothelial growth factor (VEGF), matrix metalloproteinase-2 and -9 were involved in Beclin1-mediated inhibition of migration and invasion. More importantly, our data also confirmed that Beclin1inhibited TSCC xenograft growth in vivo.Conclusion: Taken together, the results indicate that autophagy regulating gene, Beclin1, may contributeto the malignant phenotypes of TSCC cells and can be a potential target for oral cancer gene therapy.

� 2014 Elsevier Ltd. All rights reserved.

Introduction

Tongue squamous cell carcinoma (TSCC) is the most commoncancers of oral cavity, which demonstrates much more aggressivebehavior [1,2]. Despite improvements in treatment, the survivalof patients with TSCC has not been significantly improved overthe past several decades. Local or regional relapse and cervicallymph node metastasis are the most prevalent causes of death in

these patients [3,4]. So, many studies have been performed toinvestigate the effects of various biological factors on the aggres-sive potentials of TSCC and unravel the molecular mechanism ofTSCC pathogenesis [5–7]. However, these processes are still poorlyunderstood. Autophagy is an orchestrated intracellular process inwhich cytoplasmic constituents are transported by autophago-somes to lysosomes for degradation, thus is essential for growthregulation, cell differentiation and cellular homeostasis [8–10].The mammalian autophagy gene Beclin1, is a coiled-coil proteininvolved in the localization of autophagic proteins to a pre-auto-phagosomal structure [11–13]. In addition, Beclin1 regulatesautophagy at different steps by binding to different Beclin1-inter-acting proteins [14–16]. Beclin1 and its associated proteins mayalso have anti-tumor activity, potentially by modulating autopha-gic cell death [17,18]. Haploid deletion of Beclin1 is frequently

984 J. Weng et al. / Oral Oncology 50 (2014) 983–990

detected in human breast and ovarian cancers [11,19]. But, the roleof Beclin1 is still not clear in the progression of TSCC. In our previ-ous study, we confirmed that down-regulation of Beclin1 is a fre-quent event in TSCC, and its decreased expression was associatedwith T stage, clinical stage and differentiation [20]. In this study,we further elucidated the role of Beclin1 and then investigatedits potential mechanisms in TSCC progression.

Materials and methods

Cell lines and cell culture

Human tongue squamous cell carcinoma cell lines (SCC-9 andSCC-15) were maintained in Dulbecco’s modified Eagles medium(DMEM, Gibco) containing 10% fetal bovine serum (Gibco), penicil-lin (100 units/ml), and streptomycin (100 lg/ml) at 37 �C with 5%CO2 in a humidified air atmosphere containing.

Plasmids and transfection

Lentiviral shRNAs were used to knockdown Beclin1 expressionin SCC-9 and SCC-15 cells. Cells were incubated with lentivirusparticles and polybrene 8 lg/ml for 16 h and washed with med-ium. Then, cells were selected with 0.5 lg/ml puromycin for2 weeks and qPCR and western-blot were performed to confirmthe expression of Beclin1. Five different oligonucleotide sequencesof Beclin1 shRNA were tested for optimal knockdown of genes. Theselected sequences and negative control sequences were shown inSupplementary Table 1. For overexpression of Beclin1, the humanBeclin1 cDNA (Open Biosystems, USA) was PCR amplified andcloned into the lentivirus vector with the CMV promoter. Then,cells were transfected with lentivirus with our without Beclin1cDNA followed by selection with puromycin (0.5 lg/ml). Wes-tern-blot and qPCR were performed to validate its expression.

Real time PCR analysis

Total RNA was isolated with the RNeasy Total RNA kit (Invitro-gen, USA). cDNA was generated with the qScript cDNA synthesis kit(Roche, Germany) according to the instructions of the manufac-turer. Quantitative gene expression was performed for Beclin1,VEGF, MMP2 and MMP9 using LightCycler 480 SYBR Green I Mas-ter Mix Reagent Kit (Roche, Germany) and the LightCycler 480Real-time System (Roche, Germany). The data were normalizedto the internal control, GAPDH to obtain DCt. The final amount ofgene of interest relative to control samples was reported by2�DDCt method.

Western blot analysis

Cellular protein were collected and lysed with RIPA buffer withPMSF (Beyotime Biotechnology, China). The lysates were centri-fuged at 15,000g for 10 min at 4 �C. Further, to analyze the secre-tion of VEGF, MMP-2, and MMP-9 in culture supernatants, cellswere incubated in serum-free medium for 24 h and then collectedand concentrated by a speed vacuum. Then, the samples were sep-arated by SDS–PAGE and transferred to the PVDF membranes. Afterprobing with primary antibodies overnight at 4 �C and secondaryantibodies for 1 h at room temperature, the bands were detectedby enhanced chemiluminescence (ECL). Primary antibodies usedare as follows: Beclin1, VEGF, IgG (Santa Cruz, USA); LC3, MMP-2,MMP-9 (Cell Signaling, USA), GAPDH (Beyotime Biotechnology,China).

Enzyme linked immunosorbent assay

To further quantify and confirm the effects of Beclin1 on VEGFand gelatinolytic MMPs observed by western blotting, we carriedout an enzyme linked immunosorbent assay. Cells were seededon 24-well plates to grow to sub-confluent. Then, the cells werewashed with PBS and incubated with serum-free medium for48 h. The culture medium were collected, briefly centrifuged andstored at 20 �C. VEGF, MMP-2 and MMP-9 concentrations in themedium were analyzed by commercial enzyme-linked immuno-sorbent assay (ELISA) kits (Cell Signaling, USA) according to themanufacturer’s instructions.

Cell proliferation analysis

Cells were seeded in a 96-well plate at a density of 1 � 104/wellfollowed by incubation at 37 �C for 24 h. Then, 3-(4,5-dimethylthi-azol-2-yl)-2,5-diphenyltetrazolium (Sigma, USA) was added intoeach well according to the instructions of the manufacturer andsubsequently dimethyl sulfoxide (Sigma, USA) was added followedby vortexing gently for 15 min. Absorbance was determined at492 nm with a microplate reader. Cells from each group wereadded to 5 wells and experiment was performed in triplicate.Absorbance was detected at 24–96 h after transfection, and growthcurve was delineated with absorbance as the vertical axis and timeas the horizontal axis.

Cell migration and invasion assay

The cell migration and invasion were measured using the BDBioCoat system (BD Biosciences, USA) following the manufacturer’sinstructions. In brief, cells were seeded in the upper chambers, andculture medium with 10% FBS was added to the lower chambers.For invasion assay, inserts coated with matrigel were used. After24 h incubation, cells which migrated to the reverse side of insertswere stained with DAPI and quantified.

Establishment of sub-cutaneous xenograft tumor model in nude mice

Twenty-five male BALB/c nude mice (male, 6–8-week-old), ran-domly divided into five groups, were housed in a temperature-con-trolled, pathogen-free animal facility with 12-h light and darkcycles. Approximately 107 of lentivirus-infected, mock-infectedand parental SCC-9 cells in 200 ll of sterile PBS were injected sub-cutaneously into the dorsal region to establish tumors. Tumormass (xenograft) volume was measured every week from week 1to week 5. After 7 week, mice were sacrificed by cervical disloca-tion, and tumors were harvested. The research was approved bythe Ethical Committee on Animal Research of the Sun Yat-senUniversity. All experimental procedures were performed accordingto national guidelines regarding the care and use of laboratoryanimals.

Immunohistochemistry

Immunohistochemical studies on the xenograft tissue micro-arrays were performed according to our previous protocol [20].Beclin1, ATG5, VEGF, MMP2 and MMP9 staining intensity weregraded into three categories based on the percentage of positiveTSCC cells: weak, 5–35%; moderate: 35–70%; strong, >70%.

Statistical analysis

Data were analyzed using the Statistical Package for the SocialScience (SPSS, Chicago, USA), Version 17.0. All values are presentedas mean ± standard deviation (SD). The Student t test was used for

J. Weng et al. / Oral Oncology 50 (2014) 983–990 985

paired data that were normally distributed. A p-value <0.05 wasconsidered statistically significant.

Results

Autophagy-related proteins expression in TSCC cells with or withoutLV-Beclin1 and LV-shBec treatment

To investigate the role of Beclin1 on autophagy, growth, migra-tion and invasion in SCC-9 and SCC-15 cells, we first establishedtwo model systems in vitro in which Beclin1 was overexpressedor suppressed in the SCC-9 and SCC-15 cell lines by infection withlentiviruses expressing the Beclin1 (LV-Beclin1), Beclin1 shRNA1(LV-shBec-1) and Beclin1 shRNA2 (LV-shBec-2) (Fig. 1). After viralinfection, more than 95% of the cells were GFP-positive, indicatinga high efficiency of Beclin1 or shRNA delivery. The Beclin1 expres-sion in LV-Beclin1 group was significantly higher than in SCC-9 andSCC-15 cells without transfection (p < 0.05) (Fig. 1A–C). On theother hand, the expression of Beclin1 are efficiently silenced inLV-shBec-1 and LV-shBec-2 groups as compared to normal controland non-silencing group (p < 0.05) confirmed by qRT-PCR and wes-tern blot (Fig. 1D–F). Meanwhile, the expression of autophagy-related proteins, ATG4, ATG5 and LC3II were significantlyenhanced in LV-Beclin1 group while decreased in LV-shBec-1 andLV-shBec-2 groups (Fig. 2).

Effects of Beclin1 on cell growth, migration and invasion in TSCC cells

Frequent down-regulation of Beclin1 mRNA and protein inTSCC suggested a potential anti-oncogenic role of this gene. To

Fig. 1. Lentivirus delivery of Beclin1 and Beclin1-specific shRNAs to SCC-9 and SCC-15 ce(LV-Beclin1), non-silencing shRNA (LV-shCon) and two different Beclin1-specific shRNAanalyzed by western blotting for Beclin1. (B) Densitometry analysis of western blots fromBeclin1 group was significantly higher and Becin1 shRNA specifically knockdowned BecliPCR analysis of cells using primers specific for Beclin1 or GAPDH mRNA. The mRNA expreby two different Beclin1-specific shRNA as compared with controls in SCC-9 and SCC-15least 3 times.

investigate the possible anti-proliferative effects of Beclin1in vitro, a MTT assay was performed and a cell growth curve wasgenerated. As shown in Fig. 3, the proliferation of SCC-9 and SCC-15 cells infected with LV-Beclin1 was significantly suppressed(Fig. 3A and B). In contrast, Beclin1 shRNA transfected cells showedsignificantly increased viability relative to control group (Fig. 3Aand B).

To further evaluate the function of Beclin1 on TSCC cell migra-tion and invasion, transwell chambers coated with or withoutmatrigel were utilized. We found that overexpressed Beclin1 ledto a significantly decreased migratory and invasive ability whencompared with control group (p < 0.05, Fig. 3C–F). However, cellmigration and invasion activity were dramatically enhanced whenBeclin1 expression was inhibited (p < 0.05, Fig. 3C–F).

VEGF, MMP-2 and MMP-9 were involved in Beclin1-mediatedinhibition of proliferation and invasion

To investigate the potential mechanisms of Beclin1-mediatedinhibition of proliferation and invasion in TSCC cells, we furtherdetermined the VEGF, MMP-2 and MMP-9 expression levels in cellculture medium which are the functional proteins of cell prolifer-ation and invasion. As shown in Fig. 4, The western blot revealedthat promotion of Beclin1 expression could markedly decreasethe productions of VEGF, MMP-2 and MMP-9 (p < 0.05), but silenc-ing of Beclin1 effectively elevated the expressions of VEGF, MMP-2and MMP-9 (p < 0.05) in respect to control. Similar results wereobserved in ELISA experiments (Fig. 4E–J). However, mRNA expres-sion levels of MMP2 and MMP9 had no obviously changed while

lls. (A) SCC-9 and SCC-15 cells were transfected with empty vector (LV-GFP), Beclin1(LV-shBec-1 and LV-shBec-2). Forty-eight hours after transfection, the cells weresix independent experiments, respectively, showing that Beclin1 expression in LV-

n1 expression in tongue squamous cell cancer cell lines. (C) Bar graph showing qRT-ssion level of Beclin1 is up-regulated in the LV-Beclin1 group and is down-regulatedcell lines. All data is presented as mean ± SD. All the experiments were repeated at

Fig. 2. Effect of Beclin1 on expression of autophagy-related proteins in TSCC cells. (A and B) Protein expression levels of ATG4, ATG5 and LC3 were revealed by western blotanalysis in the TSCC cells transfected with LV-GFP, LV-Beclin1, non-silencing-shRNA and Beclin1 shRNAs. (C–E) Levels of LC3, ATG4 and ATG5 protein were estimated as aratio of LC3-II, ATG4 and ATG5 correspondingly to GAPDH levels. ATG4, ATG5 and LC3II were significantly enhanced in LV-Beclin1 group while decreased in LV-shBec-1 andLV-shBec-2 groups compared with control group.

986 J. Weng et al. / Oral Oncology 50 (2014) 983–990

cells were treated with Lv-shBec1 and Lv-shBec2 or Lv-Beclin1, butVEGF mRNA expression could be regulated by Beclin1 as shown inFig. 4K–M. These data suggested that Beclin1 may regulate TSCCcell invasion through the proteolytic degradation of extracellularmatrix (ECM).

Beclin1 inhibits tumor growth in xenograft tumors in vivo

The results showed that Beclin1 could suppress tumor progres-sion in vitro, then, we further evaluated the effects of Beclin1 onthe tumorigenic phenotype and in particular its contribution totumor growth in vivo. SCC-9 cells infected with or withoutLV-Beclin1 and LV-shBec-1 were injected into mice. The resultsdemonstrated that xenografts with lower expression levels ofBeclin1 grew faster than those transfected with empty vector con-trol. Conversely, overexpression of Beclin1 in SCC-9 cell line inhib-its tumor growth in vivo as compared to control group (p < 0.05)(Fig. 5). On day 35, tumors were removed and immunohistochem-istry was performed to confirm the expression levels of Beclin1,ATG5, VEGF, MMP2 and MMP9 in xenograft. As shown in Fig. 5D,Beclin1 and ATG5 staining intensity is strong in the LV-Beclin1group while VEGF, MMP2 and MMP9 staining is weak, the converseresults were observed in the Beclin1-knockdown group.

Discussion

Autophagy is a genetically programmed process which cellremoves damaged proteins and organelles to limits their cumula-tive deleterious effects [8]. Emerging evidences suggested thatderegulation of autophagy contributes to the pathogenesis of dif-ferent disease states, including neurodegenerative disorders, car-diomyopathy, skeletal myopathies, and infectious diseases [21].Moreover, most of studies indicated that the suppression ofautophagy activity was frequently correlated with malignant pro-

gress raising the possibility that defects in cellular autophagy con-tribute to the development of cancer. Specifically, malignant cellsoften display lower basal autophagic activity than their normalcounterparts in many types of tumor, such as breast cancer [22],ovarian cancer [23] and gastric cancer [24]. In rat liver carcinogen-esis models, autophagic activity is decreased slightly at a pre-neo-plastic stage and becomes more substantially diminished at a laterstage [25]. Similar results were also observed in our previous study[20]. However, it remains unknown whether the decrease in auto-phagic activity observed in malignant cells is mechanisticallyimportant or merely an epiphenomenon in malignant progress.Fortunately, the highly conserved autophagy-related genes providethe opportunity to use genetic approaches to investigate the role ofautophagy in the development of cancer [9,21,26]. Beclin1, the firstdownstream autophagy-execution gene, is a critical component inthe class III PI3 kinase complex (PI3KC3) that induces the forma-tion of autophagosomes [12,26]. Our previous studies confirm thatBeclin1 expression is dramatically decreased in TSCC which is con-sistent with other studies. To further elucidate the cellular effectsof Beclin1 in TSCC cells, stable cell lines with overexpression andsilence of Beclin1 were established and confirmed by qPCR andwestern blot. Then, cell proliferation was determined by MTTassay. We showed that Beclin1 could significantly suppress prolif-eration in TSCC cell lines and tumor growth in vivo. This is consis-tent with several previous reports that over-expression of Beclin1inhibits cell proliferation in cervical [27] and lung [28] cancer cells.However, Beclin1 also play a protective role for cancer cells whentreated with cisplatin (shown in Supplementary Fig. 1). These find-ings suggested Beclin1 has distinct role in cancer cells dependenton the stress which is in agreement with the physical role ofautophagy maintaining the cellular homeostasis.

Invasion and metastasis are the hallmarks of cancer and themain causes of death in patients with advanced TSCC [29]. Activecell migration is a critical step in the invasion and metastasis cas-cade of cancers. Since Beclin1 has been shown to regulate the

LV-Beclin 1 LV-shBec-2LV-shConSCC-9LV-GFP LV-shBec-1

LV-Beclin 1 LV-shBec-2LV-shConSCC-15LV-GFP LV-shBec-1

LV-Beclin 1 LV-shBec-2LV-shConSCC-9LV-GFP LV-shBec-1

LV-Beclin 1 LV-shBec-2LV-shConSCC-15LV-GFP LV-shBec-1

DC

A

**

*

*

**

*

**

*

*

*

FE

B

Fig. 3. Effects of Beclin1 on TSCC cells proliferation, invasion and migration. (A) Proliferation curves of SCC-9 and SCC-15 cell lines following transfection with recombinantswere determined using MTT assay. Values are normalized against the non-transfected control. (B and C) The invasion and migration ability of SCC-9 and SCC-15 cells wasevaluated by transwell chambers assay. The migrated cells were stained with DAPI and counted under bright-field microscopy at 100 magnifications. Representative photosand quantitative data are shown. Data were mean ± SD values from three experiments, each performed in triplicate. Scale bars represent 50 lm.

J. Weng et al. / Oral Oncology 50 (2014) 983–990 987

activity of autophagy through its interactions with Vps34 and PI3kinase in an mTor independent pathway [14,15], we suspected thatBeclin1 may suppress cell migration and invasion. In the presentstudy, the effects of Beclin1 on the abilities of TSCC cells to invadeand migrate were also investigated by transwell assay with orwithout matrigel. The data demonstrated that Beclin1 could inhibitmigration and invasion of TSCC cells. These results are consistentwith the role of Beclin1 in other cancer types, in which increasedBeclin1 expression impairs cell migration of cervical [27] andbreast [30] cancers indicating that Beclin1 may regulate metastasisassociated genes. To further confirm the role of autophagy in inva-sion, cancer cells were also treated with ATG5 siRNA, the resultsdemonstrated that silence of ATG5 leads to enhanced migrationof cancer cells which is in agreement with Beclin1 silence (shownin Supplementary Fig. 2). The degradation or destruction of extra-cellular matrix (ECM) and basement membrane is the crucial stepsin metastasis of cancers in which matrix metalloproteinases(MMPs) play a major role. Moreover, MMPs are also essential toeach step of capillary formation and formation of vascular endo-thelial basement membrane [31]. Angiogenesis is the process offorming new blood vessels and requires degradation of the vascu-lar basement membrane and remodeling of the ECM in order toallow cancer cells to migrate and invade into the surrounding tis-sue [32]. Several studies demonstrated MMPs is involved in the

angiogenesis through regulation of vascular endothelial growthfactor (VEGF) [31,33]. Moreover, studies demonstrated that stimu-lation of autophagy signals can inhibit angiogenesis indicating thatautophagy response could be used as a potential strategy to inhibitgrowth of new blood vessels in cancers [34]. Therefore, we furtherdetected the expression of secreted MMPs and VEGF after Beclin1overexpression and silence. Our results showed that MMP2,MMP9 and VEGF were involved in Beclin1-mediated inhibition ofinvasion in TSCC cells. To further investigate the effect of Beclin1on expression of VEGF, MMP2 and MMP9, qPCR was also per-formed to detect the mRNA expression level of VEGF, MMP2 andMMP9. However, Beclin1 seems no obviously effect on MMP2and MMP9 mRNA expression levels. These data suggested thatBeclin1 may suppress the mRNA expression of VEGF at transcrip-tional or post-transcriptional level but not only a by-stander effectat protein level. For the MMP2 and MMP9, qPCR results could notexclude the possible by-stander effect. We think a further studywill be necessary to clarify the exact mechanism in future.

Taken together, these findings indicated that Beclin1 served astumor suppressor in TSCC development. Interestingly, Tang et al.showed that LC3, ATG9a, Beclin1 and ATG5 overexpression wereassociated with a poor prognosis in patients with oral squamouscell carcinoma (OSCC) in a series of studies. These findings indi-cated increased expression of Beclin1 also has a crucial role in

Fig. 4. Effect of Beclin1 on VEGF, MMP-2 and MMP-9 expression. (A and B) The productions of VEGF, MMP-2 and MMP-9 in the culture media were analyzed using westernblot and immunosorbent assay. (C and D) Levels of VEGF, MMP-2 and MMP-9 protein were estimated as a ratio to levels of GAPDH. (E–J) The graphs show the enzyme amountof individual experiments and group means of SCC-9 and SCC-15 cells. (K–M) mRNA expression levels of VEGF, MMP2 and MMP9 were quantified by qPCR.

988 J. Weng et al. / Oral Oncology 50 (2014) 983–990

OSCC progression [35–37]. More importantly, we also found therewere distinct expression pattern of Beclin1 in different TSCC tissuesamples and different area at same samples, which demonstratednormal-like expression pattern, increased expression pattern anddecreased pattern (shown in Supplementary Fig. 3). Similar resultswere observed in colorectal cancer reported by Koukourakis et al.[38]. Therefore, further studies are necessary to clarify the role ofautophagy and Beclin1 in OSCC development by classified the

patients into three groups at least, including normal-like expres-sion pattern, increased expression pattern and decreased expres-sion pattern. We supposed that the loss of Beclin1 expressiondefines poor prognosis by modulating cancer cell migration andgrowth as shown in this manuscript, while excessive overexpres-sion of Beclin1 also defines subgroups of tumors with aggressiveclinical behavior through protecting cancer cell against stressinduced by hypoxia, acidity and chemotherapy to facilitate

A

C

B

LV-Beclin 1

LV-shBec-1

LV-GFP

SCC-9

LV-shCon

D

Fig. 5. Effect of Beclin1 on TSCC xenograft in nude mice. SCC-9 cells were transfected with GFP, Beclin1, non-silencing-shRNA and Beclin1 shRNA1 and nude mice wereinoculated subcutaneously with 1 � 107 cells at one site per mouse. The tumor mass (xenograft) volume was measured every week from week 1 to week 5. Data are expressedas the (means ± SD) and represent five independent experiments. (A) Photograph of xenografts dissected from nude mice after 5 weeks subcutaneous inoculation. (B) Tumorgrowth curve showing a significant growth tendency in Beclin1 shRNA1 group (p < 0.05). Conversely, tumor growth was delayed significantly in Beclin1 transfected group ascompared to the control group (p < 0.05). (C) Tumor weights were measured at day 35 and similar results were observed. (D) Representative images of Beclin1, ATG5, VEGF,MMP2 and MMP9 immunostaining in SCC-9 xenograft tumors. Scale bars represent 50 lm.

J. Weng et al. / Oral Oncology 50 (2014) 983–990 989

metastasis and therapy resistant. Collectively, we conclude that theroles of Beclin1 and autophagy are dependent on cancer type,stage, genetic background and microenvironment. Our findings

provide the first implication that Beclin1 is important for cellproliferation, and regulates cell invasion through modulating theproduction of VEGF, MMP-2 and MMP-9 in TSCC cells.

990 J. Weng et al. / Oral Oncology 50 (2014) 983–990

Conflict of Interest Statement

None declared.

Acknowledgements

This work was supported in part by the National NaturalScience Foundation of China (81202136), Specialized ResearchFund for the Doctoral Program of Higher Education(20120171120068), Guangdong Province Nature Science Founda-tion (10151008901000025), Guangdong Province DevelopingProgram (2009B050700024), Guangzhou Developing Program(2012J5100008) and Fundamental Research Funds for the CentralUniversities (11ykpy47).

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.oraloncology.2014.06.020.

References

[1] Rodrigues VC, Moss SM, Tuomainen H. Oral cancer in the UK: to screen or notto screen. Oral Oncol 1998;34:454–65.

[2] Greenlee RT, Hill-Harmon MB, Murray T, Thun M. Cancer statistics, 2001. CACancer J Clin 2001;51:15–36.

[3] Layland MK, Sessions DG, Lenox J. The influence of lymph node metastasis inthe treatment of squamous cell carcinoma of the oral cavity, oropharynx,larynx, and hypopharynx: N0 versus N+. Laryngoscope 2005;115:629–39.

[4] Sano D, Myers JN. Metastasis of squamous cell carcinoma of the oral tongue.Cancer Metastasis Rev 2007;26:645–62.

[5] Kurokawa H, Zhang M, Matsumoto S, Yamashita Y, Tomoyose T, Tanaka T, et al.The high prognostic value of the histologic grade at the deep invasive front oftongue squamous cell carcinoma. J Oral Pathol Med 2005;34:329–33.

[6] Bello IO, Vilen ST, Niinimaa A, Kantola S, Soini Y, Salo T. Expression of claudins1, 4, 5, and 7 and occludin, and relationship with prognosis in squamous cellcarcinoma of the tongue. Hum Pathol 2008;39:1212–20.

[7] Wu RQ, Zhao XF, Wang ZY, Zhou M, Chen QM. Novel molecular events in oralcarcinogenesis via integrative approaches. J Dent Res 2011;90:561–72.

[8] Yoshimori T. Autophagy: a regulated bulk degradation process inside cells.Biochem Biophys Res Commun 2004;313:453–8.

[9] Levine B, Klionsky DJ. Development by self-digestion: molecular mechanismsand biological functions of autophagy. Dev Cell 2004;6:463–77.

[10] Shintani T, Klionsky DJ. Autophagy in health and disease: a double-edgedsword. Science 2004;306:990–5.

[11] Liang XH, Jackson S, Seaman M, Brown K, Kempkes B, Hibshoosh H, et al.Induction of autophagy and inhibition of tumorigenesis by beclin 1. Nature1999;402:672–6.

[12] Tassa A, Roux MP, Attaix D, Bechet DM. Class III phosphoinositide 3-kinase–Beclin1 complex mediates the amino acid-dependent regulation of autophagyin C2C12 myotubes. Biochem J 2003;376:577–86.

[13] Furuya N, Yu J, Byfield M, Pattingre S, Levine B. The evolutionarily conserveddomain of Beclin 1 is required for Vps34 binding, autophagy and tumorsuppressor function. Autophagy 2005;1:46–52.

[14] Kihara A, Kabeya Y, Ohsumi Y, Yoshimori T. Beclin-phosphatidylinositol3-kinase complex functions at the trans-Golgi network. Embo Rep2001;2:330–5.

[15] Liang C, Feng P, Ku B, Dotan I, Canaani D, Oh BH, et al. Autophagic and tumoursuppressor activity of a novel Beclin1-binding protein UVRAG. Nat Cell Biol2006;8:688–99.

[16] Matsunaga K, Saitoh T, Tabata K, Omori H, Satoh T, Kurotori N, et al. Two Beclin1-binding proteins, Atg14L and Rubicon, reciprocally regulate autophagy atdifferent stages. Nat Cell Biol 2009;11:385–96.

[17] Yue Z, Jin S, Yang C, Levine AJ, Heintz N. Beclin 1, an autophagy gene essentialfor early embryonic development, is a haploinsufficient tumor suppressor.Proc Natl Acad Sci USA 2003;100:15077–82.

[18] Qu X, Yu J, Bhagat G, Furuya N, Hibshoosh H, Troxel A, et al. Promotion oftumorigenesis by heterozygous disruption of the beclin 1 autophagy gene. JClin Invest 2003;112:1809–20.

[19] Aita VM, Liang XH, Murty VV, Pincus DL, Yu W, Cayanis E, et al. Cloning andgenomic organization of beclin 1, a candidate tumor suppressor gene onchromosome 17q21. Genomics 1999;59:59–65.

[20] Wang Y, Wang C, Tang H, Wang M, Weng J, Liu X, et al. Decrease of autophagyactivity promotes malignant progression of tongue squamous cell carcinoma. JOral Pathol Med 2013;42:557–64.

[21] Klionsky DJ, Cregg JM, Dunn WJ, Emr SD, Sakai Y, Sandoval IV, et al.A unified nomenclature for yeast autophagy-related genes. Dev Cell2003;5:539–45.

[22] Won KY, Kim GY, Kim YW, Song JY, Lim SJ. Clinicopathologic correlation ofbeclin-1 and bcl-2 expression in human breast cancer. Hum Pathol2010;41:107–12.

[23] Shen Y, Li DD, Wang LL, Deng R, Zhu XF. Decreased expression of autophagy-related proteins in malignant epithelial ovarian cancer. Autophagy2008;4:1067–8.

[24] Chen YB, Hou JH, Feng XY, Chen S, Zhou ZW, Zhang XS, et al.Decreased expression of Beclin 1 correlates with a metastatic phenotypicfeature and adverse prognosis of gastric carcinomas. J Surg Oncol 2012;105:542–7.

[25] Kisen GO, Tessitore L, Costelli P, Gordon PB, Schwarze PE, Baccino FM, et al.Reduced autophagic activity in primary rat hepatocellular carcinoma andascites hepatoma cells. Carcinogenesis 1993;14:2501–5.

[26] Mizushima N. Autophagy: process and function. Genes Dev 2007;21:2861–73.[27] Sun Y, Liu JH, Sui YX, Jin L, Yang Y, Lin SM, et al. Beclin1 overexpression inhibits

proliferation, invasion and migration of CaSki cervical cancer cells. Asian Pac JCancer Prev 2011;12:1269–73.

[28] Chang SH, Minai-Tehrani A, Shin JY, Park S, Kim JE, Yu KN, et al. Beclin1-induced autophagy abrogates radioresistance of lung cancer cells bysuppressing osteopontin. J Radiat Res 2012;53:422–32.

[29] Cohen EE, Davis DW, Karrison TG, Seiwert TY, Wong SJ, Nattam S, et al.Erlotinib and bevacizumab in patients with recurrent or metastatic squamous-cell carcinoma of the head and neck: a phase I/II study. Lancet Oncol2009;10:247–57.

[30] Liang XH, Yu J, Brown K, Levine B. Beclin 1 contains a leucine-rich nuclearexport signal that is required for its autophagy and tumor suppressor function.Cancer Res 2001;61:3443–9.

[31] Ghajar CM, George SC, Putnam AJ. Matrix metalloproteinase control ofcapillary morphogenesis. Crit Rev Eukaryot Gene Expr 2008;18:251–78.

[32] Rundhaug JE. Matrix metalloproteinases and angiogenesis. J Cell Mol Med2005;9:267–85.

[33] Kim SH, Cho NH, Kim K, Lee JS, Koo BS, Kim JH, et al. Correlations of oral tonguecancer invasion with matrix metalloproteinases (MMPs) and vascularendothelial growth factor (VEGF) expression. J Surg Oncol 2006;93:330–7.

[34] Shinohara ET, Cao C, Niermann K, Mu Y, Zeng F, Hallahan DE, et al. Enhancedradiation damage of tumor vasculature by mTOR inhibitors. Oncogene2005;24:5414–22.

[35] Tang JY, Fang YY, Hsi E, Huang YC, Hsu NC, Yang WC, et al. Immunopositivity ofBeclin-1 and ATG5 as indicators of survival and disease recurrence in oralsquamous cell carcinoma. Anticancer Res 2013;33:5611–6.

[36] Tang JY, Hsi E, Huang YC, Hsu NC, Chu PY, Chai CY. High LC3 expressioncorrelates with poor survival in patients with oral squamous cell carcinoma.Hum Pathol 2013;44:2558–62.

[37] Tang JY, Hsi E, Huang YC, Hsu NC, Chen YK, Chu PY, et al. ATG9Aoverexpression is associated with disease recurrence and poor survival inpatients with oral squamous cell carcinoma. Virchows Arch 2013;463:737–42.

[38] Koukourakis MI1, Giatromanolaki A, Sivridis E, Pitiakoudis M, Gatter KC, HarrisAL. Beclin 1 over- and under expression in colorectal cancer: distinct patternsrelate to prognosis and tumour hypoxia. Br J Cancer 2010;103:1209–14.