Notch Signaling Activation Is Associated with Patient Mortality and Increased FGF1-Mediated Invasion in Squamous Cell Carcinoma of the Oral Cavity Alice N

Published OnlineFirst June 27, 2016; DOI: 10.1158/1541-7786.MCR-16-0114

Signal Transduction Molecular Cancer Research Notch Signaling Activation Is Associated with Patient Mortality and Increased FGF1-Mediated Invasion in Squamous Cell Carcinoma of the Oral Cavity Alice N. Weaver1, M. Benjamin Burch2,Tifﬁny S. Cooper1, Deborah L. Della Manna1, Shi Wei3, Akinyemi I. Ojesina4, Eben L. Rosenthal5, and Eddy S. Yang1

Abstract

Oral squamous cell carcinoma (OSCC) is a cancer subtype Notch pathway upregulation was significantly correlated with that lacks validated prognostic and therapeutic biomarkers, and patient mortality status and with expression of the proinvasive human papillomavirus status has not proven beneficial in gene FGF1. In vitro Notch activation in HNSCC cells increased predicting patient outcomes. A gene expression pathway anal- transcription of FGF1 and induced a marked increase in cell ysis was conducted using OSCC patient specimens to identify migration and invasion, which was fully abrogated by FGF1 molecular targets that may improve management of this dis- knockdown. These results reveal that increased Notch pathway ease. RNA was isolated from 19 OSCCs treated surgically at the signaling plays a role in cancer progression and patient out- University of Alabama at Birmingham (UAB; Birmingham, AL) comes in OSCC. Accordingly, the Notch–FGF interaction and evaluated using the NanoString nCounter system. Results should be further studied as a prognostic biomarker and were confirmed using the oral cavity subdivision of the Head potential therapeutic target for OSCC. and Neck Squamous Cell Carcinoma Cancer (HNSCC) study generated by The Cancer Genome Atlas (TCGA) Research Implications: Patients with squamous cell carcinoma of the Network. Further characterization of the in vitro phenotype oral cavity who succumb to their disease are more likely to produced by Notch pathway activation in HNSCC cell lines have upregulated Notch signaling, which may mediate a more included gene expression, proliferation, cell cycle, migration, invasive phenotype through increased FGF1 transcription. Mol invasion, and radiosensitivity. In both UAB and TCGA samples, Cancer Res; 14(9); 883–91. 2016 AACR.

Introduction decreased recurrence and improved overall survival in this subtype. In contrast, relapse is still common in patients with OSCC Head and neck squamous cell carcinoma (HNSCC) is a mul- and survival remains below 70% (1, 2). One reason for this tifactorial and heterogeneous disease encompassing multiple limited progress may be the lack of reliable biomarkers for OSCCs. anatomic sites but most commonly arising from the oral cavity HPV is not associated with therapeutic sensitivity or outcomes in [oral squamous cell carcinoma (OSCC)] or oropharynx tumors from the oral cavity, and therapies targeting EGFR, which [oropharyngeal squamous cell carcinoma (OPSCC)]. The major- is commonly activated in this disease, produce only a modest ity of HNSCC patients present with locally advanced disease, response (3, 4). In the absence of effective treatments, there is a requiring intense multimodality therapy associated with signifi- strong need to identify molecular targets and predictive markers to cant acute and long-term toxicities. Increasing incidence of improve the management of OSCC. human papillomavirus (HPV) in OPSCCs has corresponded with Recently, the Notch signaling pathway has emerged as a candidate to fill this void. Notch signaling plays an important role in 1Department of Radiation Oncology, University of Alabama at Bir- cell proliferation, differentiation, and apoptosis during embryo- mingham, Birmingham, Alabama. 2Department of Otolaryngology- genesis (5, 6). The canonical pathway is activated by the binding Head and Neck Surgery, University of Alabama at Birmingham, Bir- 3 of Delta or Jag ligands expressed on the cell surface to Notch mingham, Alabama. Department of Pathology, University of Alabama trans at Birmingham, Birmingham, Alabama. 4Department of Epidemiology, receptors located on the surface of a neighboring cell. This University of Alabama at Birmingham, Birmingham, Alabama. interaction induces g-secretase–mediated cleavage of the Notch 5Department of Otolaryngology-Head and Neck Surgery, Stanford intracellular domain (NICD), which translocates into the nucleus University, Palo Alto, California. (7). NICD complexes with and activates CBF1-SU(H)-LAG1 Note: Supplementary data for this article are available at Molecular Cancer (CSL) and recruits other coactivators such as Mastermind-like Research Online (http://mcr.aacrjournals.org/). (MAML) proteins (8). The resulting complex activates transcrip- Corresponding Author: Eddy S. Yang, University of Alabama at Birmingham, 619 tion of Notch target genes, including HEY1/2, HES1/5, NFkB, 19th Street South, Birmingham, AL 35249. Phone: 205-934-2762; Fax: 205-975- MYC, CCND1, BCL2, and CCR7, in a cell type-specific manner. 0784; E-mail: [email protected] Aberrant Notch signaling has been found in a number of solid doi: 10.1158/1541-7786.MCR-16-0114 and hematologic malignancies and can function in either an 2016 American Association for Cancer Research. oncogenic or tumor-suppressive capacity depending on cell type

www.aacrjournals.org 883

Weaver et al.

and context (9). For example, Notch signaling activates transcrip- Table 1. Patient demographics Cip1 Kip1 tion of cell-cycle regulators p21 and p27 in epithelial cells OSCC OSCC but not in other tissue types, suggesting the Notch pathway may patients patients w/o CSM w/ CSM function as a tumor suppressor in squamous cell carcinomas Characteristics N ¼ 14 % N ¼ 5% P (10, 11). In support of this theory, two separate exome sequencing Gender studies of HNSCC revealed 10% to 15% of tumors had inactivat- M 11 79 4 80 1.00 ing mutations in NOTCH1 (12, 13). However, this model has F321120 been challenged by evidence that HNSCCs display a bimodal Average age, y 60.57 64.80 0.61 pattern of Notch alterations, including a subset of tumors with Site of origin Mandibular alveolar ridge 2 14 0 0 0.90 increased expression of ligands JAG1 and JAG2, as well as the NOTCH1 NOTCH3 – in vitro Gingivobuccal sulcus 1 7 0 0 and receptors (14 18). In addition, Floor of mouth 5 36 2 40 work has shown a relationship between Notch1 activity and Buccal mucosa 1 7 0 0 clinically relevant characteristics, such as cancer stem cell–like Mandible 2 14 1 20 properties and TNFa-mediated inflammation (15, 16, 18). Floor of mouth þ Mandible 2 14 1 20 HNSCC candidate biomarker studies typically focus on iden- Unknown 1 7 1 20 Nodal disease tifying individual genes or proteins that correlated with patient Yes 13 93 4 80 1.00 outcomes and tumor biology. We hypothesized that dysregula- No 1 7 1 20 tion at the pathway level may significantly alter signaling activity, Average size, cm 3.67 3.60 0.92 tumor behavior, and patient outcomes, which may not be detect- T stage (AJCC) able in individual biomarker studies. Thus, we conducted a gene T2 8 57 0 0 0.04 expression pathway analysis to identify cell signaling abnormal- T4 6 43 5 100 Bone invasion ities associated with cancer-specific mortality in OSCC. We report fi Yes 6 43 4 80 0.30 asigni cant association between Notch pathway upregulation No 8 57 1 20 and mortality in two independent OSCC datasets. A positive Perineural invasion association between Notch activation and upregulation of the Yes 1 7 2 40 0.15 proinvasive gene FGF1 wasalsoobservedinOSCCsamples No 13 93 3 60 from the University of Alabama at Birmingham (UAB; Birming- Margins positive Yes 5 36 1 20 0.61 ham, AL) and validated using The Cancer Genome Atlas In vitro No 9 64 5 80 (TCGA) dataset. , activation of Notch signaling in Adjuvant XRT human head and neck cancer cells altered the expression of Yes 7 50 4 80 0.34 Notch pathway genes in a pattern comparable with that No 7 50 1 20 observed in patients with a high Notch pathway score and Average follow-up (mo) 27.09 28.44 0.92 produced a similar increase in transcription of FGF1.These Average PFS (mo) 26.44 25.44 0.94 Average OS (mo) 60.92 37.98 0.11 in vitro gene expression changes were accompanied by significantly increased cell migration and invasion in Notch-activat- Abbreviations: AJCC, American Joint Committee on Cancer; F, female; M, male; FGF1 OS, overall survival; PFS, progression free survival; w/, with; w/o, without; XRT, ed cells, which was reversed upon knockdown. Our radiotherapy. results suggest that upregulation of the Notch pathway may be a marker of invasive phenotype in OSCC.

Materials and Methods for 18 hours at 4 C prior to plating. FGF1 expression was inhibited by transfection with FGF1 siRNA (Santa Cruz Biotechnology, sc- Human tissue samples 39444) using Lipofectamine RNAiMAX (Invitrogen, REF 13778- Tissue samples were obtained from the UAB surgical pathology 150) as per manufacturer's instructions. Control siRNA-A (Santa archives (2000–2009) with Institutional Review Board approval Cruz Biotechnology, sc-37007) was used for comparison. Cells and represent histologically confirmed OSCC. All patients were were transfected for 48 hours, then trypsinized, centrifuged, and treated surgically in the Department of Otolaryngology-Head and replated for use in experiments. Where indicated, FGF1 knock- Neck Surgery at the UAB Hospital. Selected patients had previ- down was complemented by recombinant FGF1 (R&D Systems, ously untreated primary disease. Formalin-fixed paraffin-embed- 232-FA-025), which was added to growth media at 100 ng/mL. ded tissue samples from a cohort of 19 primary OSCC tumors were used (Table 1). RNA isolation and NanoString PanCancer Pathways analysis For patient samples, RNA was harvested from areas of >70% Cell culture and reagents tumor, as identified by a pathologist, using the High Pure FFPET HNSCC cell lines UM-SCC1 and UM-SCC6 were obtained RNA Isolation Kit (Roche, REF 06 650 775 001) as per the courtesy of Thomas E. Carey (University of Michigan, Ann Arbor, manufacturer's instructions. For cell lines, RNA was harvested MI; 1993). The HNSCC cell line FaDu (HTB-43) was obtained using the PureLink RNA Mini Kit (Ambion, no. 12183018A). RNA from ATCC (2011). Cell lines have not been authenticated by samples had a concentration 12.5 ng/mL and an A260/280 ratio our laboratory. All cell lines were maintained in DMEM between 1.7 and 2.3 as determined by spectrophotometer. mRNA growth medium (Sigma) supplemented with 10% FBS (SAFC was analyzed by the UAB NanoString Laboratory (www.uab.edu/ Biosciences) and 1% penicillin/streptomycin (Gibco). Notch medicine/radonc/en/nanostring; ref. 19). Samples were pro- signaling was activated by coating cell culture dishes with 1.0 cessed for analysis on the NanoString nCounter Flex system using mg/mL recombinant human DLL4 (R&D Systems, 1506-D4-050) the PanCancer Pathways Plus panel as per the manufacturer's

884 Mol Cancer Res; 14(9) September 2016 Molecular Cancer Research

Notch Activation Is Associated with Mortality in OSCC

instructions (NanoString Technologies). RCC data ﬁles were qRT-PCR imported into NanoString nSolver 2.5 and further analyzed using RNA was harvested from cell lines using the PureLink RNA the PanCancer Pathways Advanced Analysis Module with the Mini Kit (Ambion, no. 12183018A). One microgram of RNA dynamic housekeeping gene selection option. was converted into cDNA using SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen, REF18080-051) as per the manufacturer's recommendations. Expression was Analysis of TCGA OSCC data quantiﬁed by qRT-PCR using 2 TaqMan Universal PCR Master The results shown here are in part based upon data generated by Mix (Applied Biosystems, REF 4304437) and primers (Thermo- the TCGA Research Network (http://cancergenome.nih.gov/) Fisher #4331182) for FGF1 (Hs01092738_m1) and GAPDH viewed in the cBioPortal for Cancer Genomics (http://www.cbio- C C (Hs02758991_g1). FGF1 expression ¼ 2 ( tFGF1 tGAPDH). portal.org/; refs. 20, 21). Gene expression results were obtained from RNASeq V2 RSEM normalized results for samples from the oral cavity anatomic subdivision of the Head and Neck Squamous Cell proliferation Cell Carcinoma study (22) with at least 12 months of survival data Cells were seeded in 24-well plates and harvested at 0, 48, and (n ¼ 71). Notch pathway gene results were multiplied by the 96 hours. Cells were washed with PBS, trypsinized, and diluted NanoString pathway score gene weights to derive a Notch path- 1:20 in isotonic saline solution (RICCA Chemical, catalog #7210- way score for TCGA samples. 5). Diluted cells were counted using a Beckman Z1 Coulter particle counter. Cell counts represent cells/mL.

Protein expression Colony Formation Assay Protein was analyzed via SDS-PAGE as described previously Clonogenic survival was determined by the colony formation (23). Cleaved Notch1 (Val1744; Cell Signaling Technology, assay as described previously (23). Briefly, cells were treated #4147) primary antibody was used at a 1:750 dilution for immu- with the indicated dose of radiation and left undisturbed for 2 noblotting. b-Actin (Santa Cruz Biotechnology, catalog #sc- weeks. Cells were then fixed and stained (25% glutaraldehyde, 47778) levels were analyzed as a loading control. Knockdown 12 mmol/L crystal violet), and the number of colonies (>50 fi fi of FGF1 was con rmed by ELISA using the Thermo Scienti c cells) was counted. Survival fraction was calculated using the fi Pierce Human FGF1 ELISA Kit (Thermo Scienti c, EHFGF1) as per following formula: the manufacturer's instructions. ð#colonies counted in experimental plate=# cells seeded in experimental plateÞ Survival fraction ¼ Scratch assay ð#colonies counted in control Cells were plated at high density in 6-well plates and reached plate=# cells seeded in control plateÞ monolayer confluence within 10 hours. A 200-mL pipette tip was used to create a scratch. Cells were washed twice with PBS and then Experiments were performed at least in triplicate. covered with normal DMEM media. Scratches were imaged at 0, 6, 12, and 24 hours with an EVOSfl digital inverted microscope using Cell Cycle Analysis fi the transmitted light application at 2 magni cation. Wound HNSCC cells were plated in 6-well plates and incubated for 24, ¼ closure % (scratch width in pixels at time point/average scratch 48, or 72 hours. At the indicated time point, cells were washed width in pixels at time 0) 100. with PBS, trypsinized, and centrifuged. Pellets were resuspended in 1 mL cold PBS, fixed with 4 mL cold 70% ethanol, and Transwell invasion assay incubated at 20 C for 18 hours. After centrifugation, solvent Matrigel basement membrane matrix (40 mL; Corning, catalog was removed and cells were resuspended in 300 mL PBS with 1 mL #354234) was added to 24-well cell culture inserts (Falcon, REF RNase. Samples were incubated at 37 C for 30 minutes and 353097) and solidified at 37C for 20 minutes. A total of 40 mLof labeled with 30 mL propidium iodide (in PBS at a final concen- either DLL4 or PBS vehicle was added to the top of the membrane tration of 100 mg/mL) at room temperature for 10 minutes. and incubated at 37C for another 10 minutes. HNSCC cells were Samples were then assessed for cell-cycle phase using flow cyto- washed, trypsinized, centrifuged, and resuspended at a density of metry to detect DNA content. 1 106 cells/mL in serum-free DMEM. A total of 200,000 cells were added to the top of the Matrigel-coated membrane. DMEM Statistical analysis (600 mL) with 20% FBS was added to the well below the mem- Patient demographics were assessed by t test using GraphPad brane. Cells were incubated at 37C for 24 hours. After 24 hours, Prism version 4.02 (GraphPad Software). NanoString data anal- Matrigel and media were removed from the top of the membrane ysis is described above. Pathway scores and differential gene using a cotton-tipped applicator. Inserts were dipped into distilled expression was evaluated in GraphPad by t test or Mann–Whitney water to remove remaining media, then stained in 0.5% crystal test, depending on results from the D'Agostino and Pearson violet/25% glutaraldehyde/H2O for 20 minutes. Inserts were then omnibus normality test. Cluster analysis of NanoString and RNA dipped into distilled water to remove remaining stain and sequencing (RNA-seq) data was completed using GENE-E version allowed to dry. Membranes were imaged using a Zeiss Axio 3.0.204 (http://www.broadinstitute.org/cancer/software/GENE- Observer.A1 manual inverted microscope with AxioCam MRc5 E/; Broad Institute, Inc., Cambridge, MA), followed by KEGG camera and AxioVs40 software (V4.8.2.0, 2006–2010, Carl Zeiss gene set enrichment analysis using GSEA v2.2.2 (http://www. MicroImaging GmbH) at 10 magnification. Five fields were broad.mit.edu/gsea/; Broad Institute, Inc.; ref. 24). In vitro phe- imaged per insert, and the average number of invaded cells per notypic data were assessed by two-way ANOVA with Bonferroni field was calculated. posttest in GraphPad.

www.aacrjournals.org Mol Cancer Res; 14(9) September 2016 885

Weaver et al.

Results from TCGA Head and Neck Squamous Cell Carcinoma study and identified 71 patients with at least one year of survival OSCC patient characteristics data (22). When the Notch pathway gene weights used in Nineteen patients with previously untreated, surgically resect- NanoString analysis were applied to normalized RNA-seq data able OSCC were included in this study. Patients were stratified for each sample, the resulting Notch pathway scores were into two cohorts based on disease status at the conclusion of significantly higher in patients with all-cause mortality (ACM; outcomes collection in 2013 (25). The median follow-up for all n ¼ 44) than in patients with no ACM (n ¼ 27; P ¼ 0.037; Fig. patients was 27.69 months. The first cohort consisted of 5 patients 1D). We also examined mRNA expression of individual Notch (26%) who had cancer-specific mortality (CSM) with a median genes to determine which pathway components, if any, drive overall survival of 30.55 months (range 16.68–71.78) and medi- the association between Notch pathway score and mortality. In an follow-up period of 17.42 months (range 0–70.49). The OSCC samples from UAB, we identified only four Notch second cohort consisted of 14 patients (74%) who had a status pathway genes that were significantly differentially expressed of alive with disease or no evidence of disease at their last known between the CSM and no CSM cohorts: ligand DLL1, negative follow-up (no CSM) with a median follow-up period of 28.56 regulator HDAC2, transcription target HES5, and ligand JAG1 months (range 0.30–67.04). Demographics and tumor character- (Supplementary Table S1). In OSCC samples from TCGA, we istics did not differ significantly between the two cohorts with the found a modest increase in expression of ligand JAG2 and exception of increased T stage in the CSM cohort (Table 1). receptor NOTCH3 in the ACM cohort compared with the no ACM cohort (Supplementary Table S1). The Notch signaling pathway is upregulated in OSCC patients Finally, we explored patterns in gene expression data using with mortality unsupervised hierarchical cluster and gene set enrichment anal- To investigate the relationship between regulation of signaling yses. In UAB patients, we identified two primary sample clusters pathways and OSCC patient outcomes, we first conducted a that were not driven by CSM status (Supplementary Fig. S1A). pathway analysis using the NanoString platform and the Pan- Using comparative biomarker analysis, we observed a nonsignif- Cancer Pathways Advanced Analysis Module. The TGFb and icant trend toward enrichment of Notch pathway genes in sam- Notch signaling pathways were identified as having the greatest ples from the CSM cohort with a nominal P value of 0.523 degree of differential regulation between the CSM and no CSM (Supplementary Fig. S1B). Among the most highly ranked upre- cohorts (Fig. 1A). However, the association between CSM and gulated genes in samples from the CSM group were the Notch signaling pathway score reached statistical significance only for pathway members DTX3, NOTCH3, NUMBL, HES1, MAML2, the Notch pathway (P ¼ 0.032; Fig. 1B). We further observed that NOTCH1, EP300, and NOTCH2 (Supplementary Fig. S1C). Inter- Notch pathway scores were significantly higher in the CSM cohort estingly, the Notch suppressors HDAC1 and HDAC2 were among than in the no CSM cohort, indicating pathway-level upregulation the highest ranked downregulated genes in this cohort. We did not (P ¼ 0.036; Fig. 1C). To validate the association between Notch find any gene sets significantly enriched at nominal P < 0.05 in the upregulation and mortality, we examined the OSCC subgroup UAB data. In contrast, TCGA samples did not form distinct

Figure 1. Notch pathway upregulation is associated with mortality in OSCC. Gene expression pathway analysis was performed by NanoString in UAB OSCC samples (A–C)and validated using RNA-seq data from TCGA OSCC samples (D). A, heatmap of all pathway scores. Red, low pathway score; yellow, high pathway score. Apop, apotosis; CC, cell cycle; TXmisREG, transcriptional misregulation; ChromMod, chromatin modification; HH, hedgehog. B, pathway significance plot for CSM. The global significance statistic indicates the extent of differential expression of pathway genes with respect to CSM. The x-axis indicates the statistical significance of the association between pathway score and CSM. Dotted line, P ¼ 0.05. C, comparison of Notch pathway scores in UAB OSCC samples with and without CSM using normalized NanoString transcript counts. Lines, median with interquartile range. D, comparison of Notch pathway scores in TCGA OSCC samples with and without ACM using normalized RNA-seq transcript counts. Lines, median with interquartile range. , P < 0.05.

886 Mol Cancer Res; 14(9) September 2016 Molecular Cancer Research

Notch Activation Is Associated with Mortality in OSCC

clusters, possibly indicating a high degree of heterogeneity (Sup- compared with control cells (Fig. 2B). These changes were con- plementary Fig. S2A). Comparative biomarker analysis revealed sistent across all three cell lines with the exception of UM-SCC6, significant enrichment of the lysine degradation, butanoate which lacked detectable HES5 expression, and are congruous with metabolism, glutathione metabolism, and glycerophospholipid an increase in Notch-mediated transcription. We also identified a metabolism pathways in the ACM cohort, but the Notch pathway 2-fold decrease in the expression of DLL1 and pathway regulator was neither up- nor downregulated (Supplementary Fig. S2B). On LFNG in Notch-a compared with control HNSCC cells (data not the basis of these findings, the activation of Notch signaling shown). These trends are in agreement UAB OSCC samples, in identified by NanoString pathway analysis is not the result of which the average expression of DLL1 and LFNG was substantially large changes in individual Notch genes. Instead, small alterations lower in the CSM cohort compared with the no CSM cohort (6.5- in the expression of Notch ligands, receptors, and regulatory fold and 1.5-fold difference, respectively). Collectively, the factors combine to produce a significant change in Notch signal- expression of Notch genes in HNSCC cells validated the approach ing at the pathway level. In addition, our results support a of DLL4-mediated pathway activation, including a possible previous study indicating the Pathifier algorithm for pathway negative feedback loop involving DLL1. analysis may detect more subtle changes in gene expression as Along with these classical targets of Notch signaling, we also compared with the other methods (26). identified increased expression of several genes associated with tumor progression and invasion in Notch-a compared with con- In vitro activation of Notch signaling increases expression of trol HNSCC cells, including cyclin A1 (CCNA1, 1.6-fold increase), FGF1 IL1b (2.1-fold increase), SIX homeobox 1 (SIX1, 5.5-fold The Notch pathway mediates multiple cancer-associated cell increase), and FGF1 (1.8-fold increase; Fig. 2C). To determine phenotypes through transcriptional regulation, which occurs in a which of these genes play a role in Notch-mediated patient context-specific manner via activation of the Notch transcription outcomes, we examined their expression in OSCC samples from complex, repression of CREB-mediated transcription, and cross- UAB and TCGA. Although patient mortality and Notch pathway talk with the JAK-STAT pathway (27). To identify transcriptional score were not consistently associated with expression of CCNA1, changes induced by Notch activation in HNSCCs, we exposed the IL1B,orSIX1, FGF1 emerged as a potential target of Notch HNSCC cell lines UM-SCC1, UM-SCC6, and FaDu to recombi- activation in clinical samples. In UAB and TCGA tumors with nant Notch ligand DLL4 in vitro, resulting in increased Notch detectable FGF1 expression, we observed a statistically significant receptor cleavage (Fig. 2A), an indicator of canonical Notch positive correlation between normalized FGF1 transcript activation, and subsequently measured gene expression using copy number and Notch pathway score (Fig. 2D and E). The NanoString. We first examined HES1 and HES5, two well-known Pearson r value for the correlation was 0.4734 in the UAB dataset Notch targets, as markers of signaling activity. A statistically (P ¼ 0.0203, R2 ¼ 0.2241) and 0.3286 in the TCGA dataset (P ¼ significant 1.7-fold increase in expression of HES1 and a 3.3-fold 0.0121, R2 ¼ 0.1080). Some OSCC samples, however, did not increase in HES5 was observed in Notch activated (Notch-a) express the FGF1 transcript (n ¼ 0 in UAB OSCC NanoString data;

Figure 2. Notch activation increases expression of FGF1 in vitro and in clinical samples. HNSCC cell lines were treated with the Notch ligand DLL4 to activate Notch signaling in vitro. A, representative image showing expression of the cleaved NICD in cells with and without DLL4 treatment. Blots were cropped for clarity. B, normalized NanoString transcript counts of the Notch transcription targets HES1 and HES5 in cells treated with PBS vehicle [control (Ctrl)] or DLL4 (Notch-a) for 8 hours. Each bar represents one cell line. C, normalized NanoString transcript counts of the ﬁbroblast growth factor 1 gene (FGF1) in control and Notch-a cells. D, correlation between normalized NanoString FGF1 transcript counts and Notch pathway score in UAB OSCC samples. E, correlation between normalized RNA-seq FGF1 transcript counts and Notch pathway score in TCGA OSCC samples. F, combined comparison of normalized FGF1 transcript counts in UAB and TCGA OSCC samples with and without mortality. Lines, median with interquartile range. , P < 0.01. n.s., not signiﬁcant.

www.aacrjournals.org Mol Cancer Res; 14(9) September 2016 887

Weaver et al.

Figure 3. Notch activation increases HNSCC cell migration and invasion in a FGF1-dependent manner. A, HNSCC cells treated with PBS vehicle [control (Ctrl)] or DLL4 (Notch-a) for 8 hours were assessed for migration by the scratch assay. Percent wound closure was calculated at 12 hours postscratch using scratch width measured in pixels. B, representative images of the scratch assay in control and Notch-a FaDu cells, taken at 2 magnification. C, cell invasion was determined by the Transwell invasion assay. Cells (200,000) were plated per insert and incubated for 24 hours, at which point Transwell membranes were removed, stained, and imaged. Shown is the average number of invaded cells per field at 10 magnification. D, representative images of stained Transwell membranes in control and Notch-a cells at 24 hours, taken at 10 magnification. UM-SCC1 cells were transfected with control or FGF1 siRNA for 72 hours, and knockdown of FGF1 expression was validated by qRT-PCR (E) and ELISA (F). Transfected cells were treated with PBS vehicle (Ctrl), DLL4 (Notch-a), and/or recombinant FGF1 and assessed for migration by the scratch assay (G) and cell invasion by the Transwell invasion assay (H). Shown is the mean SEM from at least three independent experiments performed in triplicate, with all treatment groups compared with control for each cell line. , P < 0.001; , P < 0.05.

n ¼ 24, 27% in TCGA OSCC RNA-seq data). In addition, FGF1 was cells, UM-SCC6 a 3.6-fold difference, and FaDu a 2.5-fold differ- not significantly associated with mortality in either the UAB or ence. Notch activation also altered cell morphology as detected TCGA datasets, although there was a trend toward increased FGF1 by light microscopy; we observed substantial spreading in invad- in the ACM TCGA cohort (Fig. 2F). These results suggest that ed cells from the Notch-a group, whereas invaded control activation of Notch signaling enhances, but may not induce, FGF1 cells remained more closely associated with membrane pores expression in OSCCs, although a relationship between upregula- (Fig. 3D). To determine the role of FGF1 in Notch-mediated tion of FGF1 and patient mortality was not evident in these data. cell migration and invasion, we used siRNA to silence FGF1 in UM-SCC1 cells, resulting in an 80% decrease in both FGF1 In vitro activation of Notch signaling promotes FGF1-mediated transcript expression (Fig. 3E) and secreted FGF1 protein levels cell migration and invasion in HNSCC cell lines (Fig. 3F), and repeated the scratch and Transwell invasion assays. As Notch activation was associated with poor patient outcomes Loss of FGF1 completely abrogated the effects of Notch activation in our clinical data and with increased expression of the proinva- on wound closure and cell invasion as compared with control sive gene FGF1 in vitro, we began our phenotypic characterization siRNA-transfected cells (Fig. 3G and H). When recombinant FGF1 by investigating cell migration and invasion as indicators of tumor protein was added to the growth media, the Notch-induced aggressiveness (28). In vitro cell migration was assessed in control migratory and invasive phenotype was restored in FGF1-silenced and Notch-a cells using the wound healing (scratch) assay (29). cells (Fig. 3G and H). These results were replicated in the FaDu cell Percent wound closure was significantly higher in Notch-a cells as line by a repeat experiment, including both FGF1 knockdown and compared with control cells for all three cell lines at 12 hours (Fig. rescue (Supplementary Fig. S3). 3A and B). Difference in wound closure ranged from approxi- In addition to migration and invasion, we further characterized mately 20% in UM-SCC1 to 40% or more in UM-SCC6 and FaDu the Notch-activated phenotype in HNSCCs with respect to cell (Fig. 3A), indicating notably increased migration in Notch-a cells. proliferation, radioresponse, and cell cycle, important cancer- Next, we used the Transwell invasion assay to measure the ability related characteristics influenced by the Notch pathway in other of cells to invade through basement membrane. Cell invasion was cancer types (30). Cell proliferation rate varied slightly between significantly increased in Notch-a compared with control cells for cell lines but did not differ between Notch-a and control groups all three cell lines at 24 hours (Fig. 3C and D). UM-SCC1 (Fig. 4A). Similarly, survival fraction following 0 to 10 Gy radi- demonstrated a 2.9-fold difference between Notch-a and control ation did not change with Notch activation, and the dose

888 Mol Cancer Res; 14(9) September 2016 Molecular Cancer Research

Notch Activation Is Associated with Mortality in OSCC

Figure 4. Notch activation does not affect cell proliferation, radiosensitivity, or cell cycle. HNSCC cells treated with PBS vehicle [control (Ctrl)] or DLL4 (Notch-a) for 8 hours were assessed for cell proliferation, radiosensitivity, and cell-cycle profile. A, cells (2,500) were plated per well and collected by trypsinization at the indicated time points. Shown is the average cell count/mL as measured by Coulter counter. B, cells were treated with increasing doses of radiation (RT), left undisturbed for 2 weeks, then fixed and stained for colony counting. Dose enhancement ratio (DER) for Notch-a was calculated at 50% survival. C–E, cells were collected at the indicated time points, fixed, and labeled with propidium iodide. DNA content was evaluated by flow cytometry. Shown is the mean SEM from at least three independent experiments performed in triplicate. enhancement ratios for Notch-a cells were not significantly HNSCCs and clarify the clinical relevance of the Notch-a pheno- altered, ranging from 1.040 to 1.132 (Fig. 4B). Finally, a com- type. In addition, our results suggest Notch mediates HNSCC parison of cell-cycle profile between Notch-a and control cells did behavior through the activity of FGF1. Accordingly, the interplay not reveal differences in any of the three HNSCC cell lines tested between Notch and FGF signaling may be a biomarker for invasive (Fig. 4C–E). Taken together, these findings indicate that Notch phenotype and mortality risk in a subset of HNSCCs. Overall, activation promotes cell migration and invasion in an FGF1- these results confirm our original hypothesis that pathway level dependent manner without affecting cell proliferation, radiosen- dysregulation is detectable in HNSCC tissue and can inform the sitivity, or cell cycle in HNSCC cells in vitro. biology of this disease. Along with differential Notch pathway expression, we also identified Notch-mediated dysregulation of FGF1 in both HNSCC Discussion cell lines and OSCC patient samples. Although FGF1 is not a In this study, we report an association between the Notch classical target of Notch transcription, ours is not the first evidence signaling pathway, patient outcomes, and molecular phenotype of a regulatory relationship between Notch and FGF signaling. In in OSCC. Expression of Notch genes and weighted Notch pathway several previous studies, the constitutive activation of canonical score were significantly increased in OSCC patients with mortality (CBF1 dependent) Notch-mediated transcription was found to as compared with those without mortality in independent data- suppress FGF1 expression, while inhibition of CBF1 and its sets from UAB and TCGA. In vitro modeling of Notch activation coactivator MAML induced FGF1 expression and transport resulted in increased expression of FGF1, and the positive rela- into the extracellular compartment (31–33). Surprisingly, we tionship between Notch activity and FGF1 expression was vali- observed evidence of the opposite, as in vitro Notch activation dated in both OSCC sample sets. Finally, activation of Notch increased FGF1 transcript levels and produced an FGF1-depen- signaling produced a significant, FGF1-dependent increase in cell dent invasive phenotype. This discrepancy may be attributable to migration and invasion. These findings, which were consistent cell type-specific or ligand-specific Notch function. Another pos- across multiple cell lines and two sets of patients, validate the sibility is that downstream Notch signaling in HNSCC cells occurs presence of Notch activation at the pathway level in a subset of through the noncanonical (CBF1 independent) pathway, which is

www.aacrjournals.org Mol Cancer Res; 14(9) September 2016 889

Weaver et al.

not well characterized but has been associated with differences in tinal cytotoxicity, which may limit their use. Additional Notch- pathway output (34). Importantly, our in vitro findings were targeted therapies, including antisense Notch, RNA interference, validated in patient samples, in which Notch pathway score and soluble receptor decoys, and dominant negative variants of tran- FGF1 expression were also positively correlated. And although we scription complex members MAML and CSL, are still in the early did not identify a direct relationship between FGF1 transcript stages of development. The excitement surrounding Notch inhi- expression and mortality in this study, which may be limited by bition should also be tempered by the lack of clarity with regard to small sample size, Koole and colleagues recently reported a Notch function in HNSCCs and other solid tumors. Although we significant association between overexpression of FGF receptor describe a role for Notch in invasion and mortality, this applies 1 protein and decreased overall and disease-free survival in only to a subset of OSCC patients. We did not detect sensitivity to HNSCC (35). Additional studies of the mechanisms by which g-secretase inhibition as a single agent in HNSCC cell lines, and Notch recognizes cell context and regulates FGF signaling in a combination treatment with radiotherapy actually decreased larger patient cohort are needed to better understand pathway radiosensitivity (data not shown). It is probable that a fine balance function in this disease. of Notch signaling is maintained in normal cells and that dysre- Another important objective of the current study was to identify gulation in either direction can be detrimental for cancer man- characteristics of cell behavior associated with increased Notch agement. For these reasons, targeting downstream Notch effec- activity in HNSCC, as multiple hallmarks of cancer have been tors, including FGF1 and FGF receptors, rather than upstream associated with Notch signaling in other models (36). Our char- signaling may be a safer and more effective therapeutic strategy for acterization of Notch activation included increased cell migration Notch-a OSCC. and invasion in vitro, as well as worsened OSCC patient outcomes. In addition, we observed structural changes in Notch-a cells, Disclosure of Potential Conflicts of Interest which could be indicative of cell membrane ruffling (Fig. 3B), E.S. Yang has received speakers bureau honoraria from and is a consultant/ a feature of actively migrating cells, although this effect was not advisory board member for NanoString. No potential conflicts of interest were further characterized in the current study. Overall, these findings disclosed by the other authors. are consistent with studies in breast and prostate cancer, in which high expression of Notch genes was associated with more aggres- Authors' Contributions sive disease, as well as in melanoma where constitutive Notch Conception and design: A.N. Weaver, E.S. Yang activation enhanced tumor progression (37–39). Our data also Development of methodology: A.N. Weaver, M.B. Burch, E.S. Yang support a previous report in HNSCC, which identified a role for Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A.N. Weaver, M.B. Burch, D.L. Della Manna, Notch in determining angiogenic potential following treatment A.I. Ojesina, E.S. Yang fi with human growth factor (40). We further identi ed FGF1 as a Analysis and interpretation of data (e.g., statistical analysis, biostatistics, possible mediator of the Notch-induced migratory and invasive computational analysis): A.N. Weaver, T.S. Cooper, S. Wei, A.I. Ojesina, phenotype in HNSCC cells. Interestingly, treatment with recom- E.L. Rosenthal, E.S. Yang binant FGF1 alone was not sufficient to increase cell migration Writing, review, and/or revision of the manuscript: A.N. Weaver, T.S. Cooper, and invasion, indicating the potential role for FGF1 in altering cell D.L. Della Manna, A.I. Ojesina, E.L. Rosenthal, E.S. Yang Administrative, technical, or material support (i.e., reporting or organizing behavior is restricted to the context of Notch activation. Again, this data, constructing databases): A.N. Weaver, M.B. Burch, T.S. Cooper, E.S. Yang may be consistent with the studies cited above, which demon- Study supervision: E.S. Yang strated that Wnt, PI3K-Akt, and MAPK signaling contributed significantly to downstream Notch effects. Although we did not Acknowledgments detect changes in these pathways in our pathway expression We would like to acknowledge Lisa Clemmons, Yolanda Hartman, and analysis, both PI3K-Akt and MAPK are activated by FGF receptor Anthony Morlandt for their assistance with procuring patient tissue and clinical signaling. Thus, the mechanism by which Notch and FGF1 com- data. bine to produce increased migration and invasion may include one of these interconnected signaling pathways. Grant Support Reports of dysregulated Notch signaling in the development This work was supported by funding from the University of Alabama at and progression of solid tumors, including HNSCC, have Birmingham Department of Radiation Oncology and the ROAR Southeast increased interest in Notch inhibition as a potential anticancer Cancer Foundation. The costs of publication of this article were defrayed in part by the payment of therapeutic strategy. Small-molecule inhibitors of g-secretase, the page charges. This article must therefore be hereby marked advertisement in enzyme responsible for cleavage of Notch receptors and down- accordance with 18 U.S.C. Section 1734 solely to indicate this fact. stream signaling, are the most clinically advanced option, having progressed to early-stage clinical trials (36). However, these agents Received April 5, 2016; revised June 15, 2016; accepted June 15, 2016; are associated with significant side effects, including gastrointes- published OnlineFirst June 27, 2016.

References 1. Pulte D, Brenner H. Changes in survival in head and neck cancers in the late 4. Egloff AM, Grandis JR. Improving response rates to EGFR-targeted thera- 20th and early 21st century: a period analysis. Oncologist 2010;15:994–1001. pies for head and neck squamous cell carcinoma: candidate predictive 2. Suh Y, Amelio I, Guerrero Urbano T, Tavassoli M. Clinical update on cancer: biomarkers and combination treatment with Src inhibitors. J Oncol molecular oncology of head and neck cancer. Cell Death Dis 2014;5:e1018. 2009;2009:896407. 3. Hansen AR, Siu LL. Epidermal growth factor receptor targeting in head and 5. Ntziachristos P, Lim Jing S, Sage J, Aifantis I. From ﬂy wings to targeted neck cancer: have we been just skimming the surface? J Clin Oncol 2013; cancer therapies: a centennial for Notch signaling. Cancer Cell 2014;25: 31:1381–3. 318–34.

890 Mol Cancer Res; 14(9) September 2016 Molecular Cancer Research

Notch Activation Is Associated with Mortality in OSCC

6. Yin L, Velazquez OC, Liu Z-J. Notch signaling: emerging molecular targets 24. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, for cancer therapy. Biochem Pharmacol 2010;80:690–701. et al. Gene set enrichment analysis: a knowledge-based approach for 7. De Strooper B, Annaert W, Cupers P, Saftig P, Craessaerts K, Mumm JS, et al. interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A A presenilin-1-dependent [gamma]-secretase-like protease mediates 2005;102:15545–50. release of Notch intracellular domain. Nature 1999;398:518–22. 25. Kejner AE, Burch MB, Sweeny L, Rosenthal EL. Bone morphogenetic 8. Wilson JJ, Kovall RA. Crystal structure of the CSL-Notch-Mastermind protein 6 expression in oral cavity squamous cell cancer is associated with ternary complex bound to DNA. Cell 2006;124:985–96. bone invasion. Laryngoscope 2013;123:3061–5. 9. Previs RA, Coleman RL, Harris AL, Sood AK. Molecular pathways: trans- 26. Drier Y, Sheffer M, Domany E. Pathway-based personalized analysis of lational and therapeutic implications of the Notch signaling pathway in cancer. Proc Natl Acad Sci U S A 2013;110:6388–93. cancer. Clin Cancer Res 2015;21:955–61. 27. Ranganathan P, Weaver KL, Capobianco AJ. Notch signalling in solid 10. Leong KG, Karsan A. Recent insights into the role of Notch signaling in tumours: a little bit of everything but not all the time. Nat Rev Cancer tumorigenesis. Blood 2006;107:2223–33. 2011;11:338–51. 11. Rangarajan A, Talora C, Okuyama R, Nicolas M, Mammucari C, Oh H, et al. 28. Li J, Wei Z, Li H, Dang Q, Zhang Z, Wang L, et al. Clinicopathological Notch signaling is a direct determinant of keratinocyte growth arrest and significance of fibroblast growth factor 1 in non–small cell lung cancer. entry into differentiation. EMBO J 2001;20:3427–36. Hum Pathol 2015;46:1821–8. 12. Agrawal N, Frederick MJ, Pickering CR, Bettegowda C, Chang K, Li RJ, et al. 29. Liang C-C, Park AY, Guan J-L. In vitro scratch assay: a convenient and Exome sequencing of head and neck squamous cell carcinoma reveals inexpensive method for analysis of cell migration in vitro. Nat Protoc inactivating mutations in NOTCH1. Science 2011;333:1154–7. 2007;2:329–33. 13. Stransky N, Egloff AM, Tward AD, Kostic AD, Cibulskis K, Sivachenko A, 30. Roy M, Pear WS, Aster JC. The multifaceted role of Notch in cancer. Curr et al. The mutational landscape of head and neck squamous cell carcinoma. Opin Gen Dev 2007;17:52–9. Science 2011;333:1157–60. 31. Kacer D, McIntire C, Kirov A, Kany E, Roth J, Liaw L, et al. 14. Sun W, Gaykalova DA, Ochs MF, Mambo E, Arnaoutakis D, Liu Y, et al. Regulation of nonclassical FGF1 release and FGF-dependent cell Activation of the NOTCH pathway in head and neck cancer. Cancer Res transformation by CBF1-mediated Notch signaling. J Cell Physiol 2014;74:1091–104. 2011;226:3064–75. 15. Yoshida R, Nagata M, Nakayama H, Niimori-Kita K, Hassan W, Tanaka T, 32. Duarte M, Kolev V, Kacer D, Mouta-Bellum C, Soldi R, Graziani et al. The pathological significance of Notch1 in oral squamous cell I, et al. Novel cross-talk between three cardiovascular regu- carcinoma. Lab Invest 2013;93:1068–81. lators thrombin cleavage fragment of Jagged1 induces fibroblast 16. Lee SH, Hong HS, Liu ZX, Kim RH, Kang MK, Park NH, et al. TNFalpha growth factor 1 expression and release. Mol Biol Cell 2008;19: enhances cancer stem cell-like phenotype via Notch-Hes1 activation in oral 4863–74. squamous cell carcinoma cells. Biochem Biophys Res Commun 2012; 33. Small D, Kovalenko D, Soldi R, Mandinova A, Kolev V, Trifonova R, et al. 424:58–64. Notch activation suppresses fibroblast growth factor-dependent cellular 17. Leethanakul C, Patel V, Gillespie J, Pallente M, Ensley JF, Koontongkaew S, transformation. J Biol Chem 2003;278:16405–13. et al. Distinct pattern of expression of differentiation and growth-related 34. Nofziger D, Miyamoto A, Lyons KM, Weinmaster G. Notch signaling genes in squamous cell carcinomas of the head and neck revealed by the use imposes two distinct blocks in the differentiation of C2C12 myoblasts. of laser capture microdissection and cDNA arrays. Oncogene 2000;19: Development 1999;126:1689–702. 3220–4. 35. Koole K, Brunen D, van Kempen PM, Noorlag R, de Bree R, Lieftink C, et al. 18. Lee SH, Do SI, Lee HJ, Kang HJ, Koo BS, Lim YC. Notch1 signaling FGFR1 is a potential prognostic biomarker and therapeutic target in head contributes to stemness in head and neck squamous cell carcinoma. Lab and neck squamous cell carcinoma. Clin Cancer Res 2016 Mar 2. [Epub Invest 2016;96:508–16. ahead of print]. 19. Sun Z, Asmann YW, Kalari KR, Bot B, Eckel-Passow JE, Baker TR, et al. 36. Koch U, Radtke F. Notch and cancer: a double-edged sword. Cell Mol Life Integrated analysis of gene expression, CpG island methylation, and gene Sci 2007;64:2746–62. copy number in breast cancer cells by deep sequencing. PLoS One 2011;6: 37. Reedijk M, Odorcic S, Chang L, Zhang H, Miller N, McCready DR, et al. e17490. High-level coexpression of JAG1 and NOTCH1 is observed in human 20. Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al. breast cancer and is associated with poor overall survival. Cancer Res Integrative analysis of complex cancer genomics and clinical profiles using 2005;65:8530–7. the cBioPortal. Sci Signal 2013;6:pl1. 38. Santagata S, Demichelis F, Riva A, Varambally S, Hofer MD, Kutok JL, et al. 21. Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, et al. The cBio JAGGED1 expression is associated with prostate cancer metastasis and cancer genomics portal: an open platform for exploring multidimensional recurrence. Cancer Res 2004;64:6854–7. cancer genomics data. Cancer Discov 2012;2:401–4. 39. Balint K, Xiao M, Pinnix CC, Soma A, Veres I, Juhasz I, et al. Activation of 22. The Cancer Genome Atlas Network. Comprehensive genomic characteri- Notch1 signaling is required for b-catenin–mediated human primary zation of head and neck squamous cell carcinomas. Nature 2015;517: melanoma progression. J Clin Invest 2005;115:3166–76. 576–82. 40. Zeng Q, Li S, Chepeha DB, Giordano TJ, Li J, Zhang H, et al. 23. Nowsheen S, Cooper T, Stanley JA, Yang ES. Synthetic lethal interactions Crosstalk between tumor and endothelial cells promotes tumor between EGFR and PARP inhibition in human triple negative breast cancer angiogenesis by MAPK activation of Notch signaling. Cancer Cell cells. PLoS One 2012;7:e46614. 2005;8:13–23.

www.aacrjournals.org Mol Cancer Res; 14(9) September 2016 891

Notch Signaling Activation Is Associated with Patient Mortality and Increased FGF1-Mediated Invasion in Squamous Cell Carcinoma of the Oral Cavity

Alice N. Weaver, M. Benjamin Burch, Tiffiny S. Cooper, et al.

Mol Cancer Res 2016;14:883-891. Published OnlineFirst June 27, 2016.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-16-0114

Supplementary Access the most recent supplemental material at: Material http://mcr.aacrjournals.org/content/suppl/2016/06/25/1541-7786.MCR-16-0114.DC1

Cited articles This article cites 39 articles, 17 of which you can access for free at: http://mcr.aacrjournals.org/content/14/9/883.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/14/9/883.full#related-urls

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/14/9/883. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.