Skip to main content
NCBI home page
The American Journal of Pathology logoLink to The American Journal of Pathology
. 2016 Apr;186(4):1055–1064. doi: 10.1016/j.ajpath.2015.11.021

Neuropilin 1 Receptor Is Up-Regulated in Dysplastic Epithelium and Oral Squamous Cell Carcinoma

Shokoufeh Shahrabi-Farahani ∗, Marina Gallottini †, Fabiana Martins †, Erik Li ∗, Dayna R Mudge ‡, Hironao Nakayama ∗,§, Kyoko Hida , Dipak Panigrahy ‡,‖, Patricia A D'Amore ∗∗,††, Diane R Bielenberg ∗,§,∗
PMCID: PMC4822338  PMID: 26877262

Abstract

Neuropilins are receptors for disparate ligands, including proangiogenic factors such as vascular endothelial growth factor and inhibitory class 3 semaphorin (SEMA3) family members. Differentiated cells in skin epithelium and cutaneous squamous cell carcinoma highly express the neuropilin-1 (NRP1) receptor. We examined the expression of NRP1 in human and mouse oral mucosa. NRP1 was significantly up-regulated in oral epithelial dysplasia and oral squamous cell carcinoma (OSCC). NRP1 receptor localized to the outer suprabasal epithelial layers in normal tongue, an expression pattern similar to the normal skin epidermis. However, dysplastic tongue epithelium and OSCC up-regulated NRP1 in basal and proliferating epithelial layers, a profile unseen in cutaneous squamous cell carcinoma. NRP1 up-regulation is observed in a mouse carcinogen-induced OSCC model and in human tongue OSCC biopsies. Human OSCC cell lines express NRP1 protein in vitro and in mouse tongue xenografts. Sites of capillary infiltration into orthotopic OSCC tumors correlate with high NRP1 expression. HSC3 xenografts, which express the highest NRP1 levels of the cell lines examined, showed massive intratumoral lymphangiogenesis. SEMA3A inhibited OSCC cell migration, suggesting that the NRP1 receptor was bioactive in OSCC. In conclusion, NRP1 is regulated in the oral epithelium and is selectively up-regulated during epithelial dysplasia. NRP1 may function as a reservoir to sequester proangiogenic ligands within the neoplastic compartment, thereby recruiting neovessels toward tumor cells.

Oral squamous cell carcinoma (OSCC) is the most malignant tumor of the oral cavity. OSCC is more aggressive than cutaneous squamous cell carcinoma (CSCC). Although the incidence of OSCC is 20 times lower than CSCC with 30,000 (0.01%) and 700,000 (0.2%) new cases each year in the United States, respectively1, 2; two-thirds of OSCC patients have evidence of disseminated disease at diagnosis,1 yet CSCC is rarely malignant with only 4% of cases developing nodal metastases.2 All tumor growth beyond the volume of 1 to 2 mm3 is angiogenesis dependent.3 The extent of tumor angiogenesis and lymphangiogenesis are two of the most important prognostic factors in OSCC.4 Understanding the mechanisms that control tumor neovascularization may lead to new therapeutic options for cancer patients.

Neuropilin 1 (NRP1) has been studied extensively in the vascular system where it acts as a coreceptor for angiogenic proteins such as vascular endothelial growth factor (VEGF) and hepatocyte growth factor (HGF) and in the neuronal system where it serves as a receptor for guidance molecules called class 3 semaphorins (SEMA3s).5, 6 Our laboratory previously demonstrated that epithelial cells in the skin and CSCC tumor cells express high levels of NRP1.7, 8, 9 However, the function of NRP1 in epithelial cells and carcinoma cells is not as well understood as its role in endothelial cells.10, 11

Several studies have reported that overexpressing the NRP1 receptor via transfection into tumor cells results in enhanced tumor size in vivo, although NRP1 does not directly increase proliferation of tumor cells in vitro.12, 13 In addition, tumors overexpressing NRP1 formed hypervascular xenografts, suggesting that NRP1 expression in the tumor compartment influences neovascularization from the stromal compartment.14, 15 This effect is apparently caused by the binding (and release) of angiogenic factors [VEGF, placenta growth factor (PlGF), HGF] to the NRP1 receptor on tumor cells, resulting in an increased chemogradient of these ligands within the local tumor microenvironment that attracts and induces sprouting of neighboring endothelial cells.

To test our hypothesis that NRP1 may increase the vascularity and invasiveness of OSCC, we investigated the expression and function of NRP1 in oral (tongue) epithelial cells and OSCC cells. Our results found an up-regulation in NRP1 receptor expression in oral dysplastic epithelium that persists throughout the stages of oral carcinogenesis and progression. In addition, the expression of NRP1 protein within the OSCC tumor compartment correlates with the pattern of blood vessel infiltration within the tumor microenvironment.

Materials and Methods

Cell Culture

HSC3, a human OSCC isolated from the cervical lymph node of a 64-year-old man with poorly differentiated SCC of the tongue, was obtained from the Japanese Cancer Research Bank (JCRB, Tokyo, Japan).16, 17 SCC9, a T2N1M0 human SCC of the tongue from a 25-year-old man with no prior treatment, and SCC68, a T4N1M0 human SCC of the tongue from a man of unknown age, were obtained from Dr. James Rheinwald (Department of Dermatology, Harvard Skin Disease Research Center, Brigham & Women's Hospital, Boston, MA).18, 19, 20 Normal oral keratinocytes were obtained from Dr. Karl Munger (Department of Pathology, Harvard Medical School, Boston, MA) and grown in keratinocyte growth medium (Lonza, Hopkinton, MA).21 Porcine aortic endothelial (PAE) cells and PAE transfected with human NRP1 were obtained from Dr. Michael Klagsbrun (Vascular Biology Program, Boston Children's Hospital, Harvard Medical School, Boston, MA) and cultured in Ham's F12 nutrient mixture.22 HSC3 was grown in Dulbecco's modified Eagle medium; SCC9 and SCC68 were grown in Dulbecco's modified Eagle medium/F12. Media were supplemented with 10% fetal bovine serum and 1% glutamine-penstrep.

Proliferation Assay

Human OSCC cells (4 × 104 cells/well) were plated in 24-well tissue culture dishes in complete media. The next day, the media were replaced with 0.2% serum-containing media with 0, 400, or 800 ng/mL SEMA3A. SEMA3A protein was purified as previously described.23 After 3 days, cells were trypsinized and resuspended in REPIPET II. The number of cells in each well was counted with a Coulter Counter. The basal proliferation rate of each cell line differed; therefore, cell counts were normalized to the percentage of each control.

Migration Assay

Human OSCC cells (104 cells) in 1% fetal bovine serum media were plated in the insert of 24-well transwell chambers with 8.0-μm pore size. Media that contained 10% fetal bovine serum were placed in the lower chambers as a chemoattractant. Cells were treated with 0, 300, or 600 ng/mL SEMA3A in the upper chamber and allowed to migrate for 8 to 10 hours at 37°C. Membranes were fixed and stained with Diff Quik. The number of migrated cells per ×200 field were counted and averaged from eight fields. Migration was plotted as the percentage of control because each cell line migrated differently.

ELISA

VEGF secreted from human OSCC (tongue) tumor cells (HSC3, SCC9, SCC68) was measured by enzyme-linked immunosorbent assay (ELISA; Human VEGF Quantikine ELISA kit; R&D Systems, Minneapolis, MN). Cells were grown to 80% confluence in 10-cm tissue culture dishes in complete media, which was replaced with 4 mL of serum-free media, and incubated overnight. A standard curve of VEGF concentration was plotted versus optical density at 450 nm. The VEGF concentration from each human cell line was extrapolated and normalized by dividing by the total protein concentration in each sample.

Western Blot Analysis

Human OSCC cells were grown to 80% confluence in 10-cm tissue culture dishes and then lyzed in protein lysis buffer (RIPA buffer plus complete protease inhibitor cocktail). Reduced proteins were run on 7.5% SDS-PAGE and transferred to nitrocellulose. Membranes were blocked with nonfat milk and incubated with either rabbit monoclonal anti-neuropilin 1 [(EPR3113); AB81321; Abcam, Cambridge, MA] or rabbit polyclonal anti–glyceraldehyde-3-phosphate dehydrogenase (AB9485; Abcam). Membranes were incubated with horseradish peroxidase-linked donkey anti-rabbit (NA934V; GE Healthcare, Chalfont St. Giles, UK) and detected with Western Lightning Plus ECL (PerkinElmer, Waltham, MA).

Human Patient Samples and Mouse OSCC Array

Institutional review board approval was obtained from the Dental School of the University of São Paulo, Brazil (2004/2010 CAAE 0031.0.017.000-10). Human specimens were retrieved from the oral pathology laboratory archives. Cases were selected from the database according to the final diagnosis as follows: epithelial dysplasia and OSCC. Hematoxylin and eosin (H&E)-stained sections of all cases were reviewed by two trained oral pathologists (F.M. and M.G.) to confirm the histopathologic diagnosis and to classify all specimens into the mild, moderate, and severe oral epithelial dysplasia categories. Supplemental Table S1 shows patient numbers and clinicopathologic markers (n = 10 to 18 patients per group; n = 61 patients total).

Paraffin-embedded mouse oral cancer tissue microarrays (no. LCOHR-MOCA) that contained tongue specimens from mice treated orally with 4-nitroquinoline-1-oxide were obtained from the Center for Oral Health Research, Medical University of South Carolina. Selected specimens ranged from normal epithelium, hyperplasia, epithelial dysplasia, papilloma, and OSCC.

Animal Studies

All mice were maintained under specific pathogen-free conditions in a facility accredited by the Association for Assessment and Accreditation of Laboratory Animal Care. The care and experimental procedures were in accordance with current regulations and were approved by the Institutional Animal Care and Use Committee at Boston Children's Hospital. VEGFLacZ mice were originally obtained from Dr. Andras Nagy (Samuel Lunenfield Institute, Toronto, ON, Canada). Adult (8-week-old) female, immunocompromised BALB/c Nude mice were purchased from Massachusetts General Hospital. Human OSCC cells (106 cells/20 μL Hanks' balanced salt solution) were injected orthotopically into the right lateral surface of the tongue. Thirty days later, mice were euthanized; tumors were removed, fixed in formalin, embedded in paraffin, and examined histopathologically.

IHC

For immunohistochemistry (IHC) formalin-fixed, paraffin-embedded sections were deparaffinized in xylene and rehydrated through a graded series of alcohols to water. Antigens were retrieved with 20 μg/mL proteinase K for 10 minutes at 37°C, endogenous peroxidase was inhibited with 3% H2O2 in methanol, and nonspecific proteins were blocked in 0.1 mol/L Tris-HCl, pH7.5; 0.15 mol/L NaCl, 0.5% blocking reagent (PerkinElmer FP1012). Sections were incubated in primary antibodies overnight at 4°C, including rabbit anti-NRP1 (C-terminus; 34-7300; Invitrogen, Carlsbad, CA/Zymed),24 rabbit monoclonal anti-Nrp1 (C-terminus; EPR3113, Abcam/Epitomics),9 Syrian hamster anti-mouse podoplanin (Reliatech, Wolfenbüttel, Germany),25 and monoclonal rat anti-mouse CD31 (MEC13.3; BD Pharmingen, San Diego, CA).9 Sections were incubated in secondary antibodies, including biotinylated goat anti-rabbit (Vector Laboratories, Burlingame, CA), biotinylated goat anti-rat (mouse absorbed; Vector Laboratories), and horseradish peroxidase-conjugated rabbit anti-Syrian hamster (Jackson ImmunoResearch Laboratories Inc., West Grove, PA). Staining was amplified with either Vectastain Elite horseradish peroxidase-linked avidin or Tyramide Signal Amplification biotin kits (PerkinElmer) and was visualized with diaminobenzidine (Vector Laboratories). Cells were counterstained with hematoxylin and mounted with Permount.

X-Gal Staining

β-galactosidase activity was detected in cryosections from VEGFLacZ mice fixed in methanol and incubated (37°C) in X-Gal reagent [1 mg/mL X-Gal (5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside) in dimethyl sulfoxide; 5 mmol/L K3Fe(CN)6; 5 mmol/L K4Fe(CN)6; 2 mmol/L MgCl2 in phosphate-buffered saline; pH 6.5]. Blue nuclei represented cells that expressed VEGF. Sections were counterstained with eosin and mounted with Permount.

Microvessel Density Analysis

Blood vessel density (CD31+ vessels/mm2) and lymphatic vessel density [LVD; podoplanin+ vessels/mm2] was counted with the use of IP Labs software version 3.5 (Scanalytics, Inc., Fairfax, VA). Tumor sections photographed at ×100 magnification (5 to 10 fields) were counted, adjusted to square millimeters, and averaged.

Statistical Analysis

Data from proliferation assays, migration assays, and microvessel density analyses were analyzed with the use of unpaired two-sample Student's t-test. Statistical significance was considered at P ≤ 0.05.

Results

Nrp1 Receptor Expression in Normal Mouse Tongue

We previously reported high Nrp1 expression in differentiated epithelium in mouse and human skin.9 Herein, we examined Nrp1 protein localization in oral epithelium in mouse paraffin sections of tongue with the use of IHC. Nrp1 receptor protein was highly expressed in all suprabasal epithelial cells of the dorsal or ventral surface of the tongue (Figure 1, A and B). Basal epithelial cells lacked Nrp1 expression (Figure 1, A and B). Normal quiescent tongue capillaries also lacked Nrp1 expression. Control sections exposed to secondary antibody alone showed no specific staining in the tongue (data not shown). H&E staining revealed the normal tongue architecture (Figure 1, C and D). Localization of the Nrp1 ligand, VEGF, was determined with the use of cryosections of tongue tissue from VEGF-LacZ transgenic mice. X-Gal staining found that VEGF was produced primarily by suprabasal epithelial cells in the dorsal and ventral tongue and in some dermal stromal cells (Figure 1, E and F).

Figure 1.

Figure 1

Open in a new tab

Nrp1 receptor and VEGF ligand are expressed in suprabasal epithelial layers of the mouse tongue. A–D: Paraffin sections of adult mouse tongue, dorsal surface (A and C), or ventral surface (B and D) were stained by immunohistochemistry with the use of anti-Nrp1 antibody (brown color) and counterstained with hematoxylin (blue color) (A and B) or H&E (C and D). Nrp1 is observed in suprabasal epithelium. The basal epithelial layer (small arrows) lacks Nrp1 expression. Quiescent capillaries in the mucosa and muscle layers do not express Nrp1. E and F: Cryosections of adult VEGFLacZ mice tongue, dorsal (E) and ventral (F) surface were stained with X-Gal reagent (blue color) and counterstained with eosin (pink color). Blue nuclei (large arrows) denote cells actively producing VEGF ligand. VEGF is expressed predominantly in suprabasal epithelial cells and submucosal stromal cells. Scale bar = 100 μm. H&E, hematoxylin and eosin; Nrp1, mouse neuropilin-1; VEGF, vascular endothelial growth factor; X-Gal, 5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside.

Nrp1 Is Up-Regulated in Mouse Oral Dysplastic Epithelium and OSCC

We next compared the expression and localization of Nrp1 protein in various stages of tumor progression in a mouse carcinogen-induced carcinoma model. Tissue microarray slides of mouse tissue harvested from animals exposed to 4-nitroquinoline-1-oxide in the drinking water were stained by IHC for mouse Nrp1. Normal tongue expressed Nrp1 receptor only in differentiated layers of the epithelium (Figure 2A); normal epithelial basal cells were devoid of Nrp1 protein. In contrast, Nrp1 was expressed in basal cells in areas of epithelial dysplasia (Figure 2, B and C). More advanced lesions showed strong Nrp1 staining in dysplastic and OSCC cells (Figure 2, D and E). In some tumors, Nrp1 expression was heterogeneous with some areas strongly expressing Nrp1 and other areas more weakly expressing Nrp1 (Figure 2E). In lighter staining areas, Nrp1 expression was apparent on the membrane of tumor cells. H&E images of representative cores and the entire tissue microarray are shown in Supplemental Figure S1.

Figure 2.

Figure 2

Open in a new tab

Nrp1 is up-regulated in carcinogen-induced dysplastic epithelium and OSCC in mouse tongue. Paraffin sections of a mouse oral cancer TMA that contains specimens from 4NQO-treated mice at various stages of tumor progression were stained by immunohistochemistry for Nrp1 (brown color). Sections were counterstained with hematoxylin (blue nuclei). Representative images are shown. A: Normal mouse tongue, ventral tip. The inset shows the dorsal surface. Note, Nrp1 is found only in differentiated, suprabasal epithelial cells (inset, arrowheads point to the lack of staining in basal cells). B and C: Epithelial dysplasia in the mouse tongue. Arrows denote areas in which the normal epithelium transitions to dysplastic epithelium and up-regulate Nrp1 expression in the basal epithelial layer. D: OSCC in the mouse tongue. Nrp1 is highly expressed in the OSCC lesion and in neighboring dysplastic epithelium. E: Nrp1 expression is heterogeneous within the OSCC. Matching H&E images and entire TMA are shown in Supplemental Figure S1. Scale bar = 100 μm. Original magnification: ×100 (main images); ×400 (inset). Nrp1, mouse neuropilin-1; OSCC, oral squamous cell carcinoma; TMA, tissue microarray; 4NQO, 4-nitroquinoline-1-oxide.

NRP1 Is Up-Regulated in Human Oral Epithelial Dysplasia

Human oral specimens were analyzed for the expression of NRP1. Human oral epithelium is much thicker than mouse oral epithelium, and degrees of dysplasia from mild to severe can be more aptly identified and described. Paraffin sections of human oral epithelial dysplasia were stained by IHC with antibodies to human NRP1 (full-length receptor) as previously described.9 Mild (Figure 3, A and B), moderate (Figure 3, C and D), and severe (Figure 3, E and F) epithelial dysplasia in human tongue samples all showed increased NRP1 receptor expression compared with normal tongue epithelium (Figure 3, G and H). NRP1 protein localized to suprabasal cells in normal tongue (Figure 3H), but expression was increased in basal cells in the areas of dysplasia (Figure 3, B, D and F). NRP1 expression and the number of cell layers involved correlated with the degree of epithelial dysplasia. Mucosal blood vessels also exhibited robust expression of NRP1 in dysplastic samples but not in the normal tongue samples.

Figure 3.

Figure 3

Open in a new tab

NRP1 is up-regulated in dysplastic epithelium in human tongue. Paraffin sections of human oral dysplasia (A–F) and human normal tongue (G and H) were stained by H&E (A, C, E, and G) and immunohistochemistry for human NRP1 (B, D, F, and H). Shown are mild epithelial dysplasia (A and B), moderate epithelial dysplasia (C and D), severe epithelial dysplasia (E and F), and normal tongue (G and H). NRP1 (brown color) is up-regulated in basal and proliferating cell layers in all degrees of dysplastic epithelium. Dermal blood vessels also up-regulate NRP1 expression in dysplastic specimens. Human normal tongue expresses NRP1 only in outer differentiated epithelial cell layers (H). Some sections were counterstained with hematoxylin (blue nuclei). Scale bar = 100 μm. H&E, hematoxylin and eosin; NRP1, human neuropilin-1.

SEMA3A Inhibition of OSCC Migration Is Proportional to NRP1 Expression

To determine whether the NRP1 receptor in OSCC tumor cells was bioactive and to examine the function of NRP1 in OSCC cells, we used an in vitro system. NRP1 protein expression was examined in cultured human OSCC cell lines by Western blot analysis. Glyceraldehyde-3-phosphate dehydrogenase was used as a loading control. All OSCC cell lines examined expressed NRP1 protein to various degrees, with HSC3 the highest and SCC68 the lowest (Figure 4A). PAE cells were used as a negative control, and PAE NRP1 cells were used as a positive control. The expression of NRP1 in all OSCC cell lines was higher than that of normal oral keratinocytes (Supplemental Figure S2L). The amount of VEGF secreted into the conditioned media by the OSCC cell lines was analyzed by ELISA and standardized to the amount of total secreted protein. HSC3 secreted the most VEGF protein, followed by SCC9 and SCC68 (Figure 4B).

Figure 4.

Figure 4

Open in a new tab

Human OSCC cell lines express a functional NRP1 receptor. A: Western blot analysis for NRP1 protein (130 kDa) in human OSCC cell lines. The membrane was stripped and reprobed for GAPDH (37 kDa) as a loading control. PAE cells were used as a negative control, and PAE NRP1 cells were used as a positive control. B: ELISA was used to determine the concentration of VEGF in picograms (pg) in the conditioned media from each OSCC cell line. Results were normalized to the protein content in each sample. C: Migration assay of OSCC cell lines incubated with SEMA3A. Migration is significantly (P < 0.05) different from control for each dose and each cell line. D: Proliferation assay (cell counts normalized to 0 dose of SEMA3A) of OSCC cell lines incubated with SEMA3A for 72 hours. SEMA3A inhibits migration but not proliferation in OSCC lines. E and F: BVD (CD31+ vessels/mm2) and LVD (Pdpn+ vessels/mm2) were determined in human OSCC tumors implanted in mouse tongue. LVD is significantly higher in HSC3 tumors compared with SCC9 or SCC68. Data are expressed as means ± SD. **P < 0.01, ****P < 0.0001. BVD, blood vessel density; ELISA, enzyme-linked immunosorbent assay; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; LVD, lymphatic vessel density; NRP1, neuropilin-1; OSCC, oral squamous cell carcinoma; PAE, porcine aortic endothelial; Pdpn, podoplanin; SEMA3A, class 3A semaphorin; VEGF, vascular endothelial growth factor.

We next analyzed whether SEMA3A, the inhibitory ligand of NRP1, could influence the migration or proliferation of OSCC cells as it does for endothelial cells.26, 27 Purified SEMA3A protein inhibited the migration of all OSCC cell lines in a dose-dependent fashion (Figure 4C). The degree of inhibition correlated with the expression level of NRP1 by each cell line. Accordingly, HSC3 cells, which expressed the highest level of NRP1, were inhibited to the highest extent. Under basal conditions, HSC3 migrated the most (average of 160 cells/field), followed by SCC9 (average of 85 cells/field), and SCC68 (average of 45 cells/field). In contrast to its effect on migration, SEMA3A did not inhibit proliferation of human OSCC cell lines (Figure 4D).

NRP1 and Microvessel Density in Human OSCC Xenografts

Three human OSCC cell lines, HSC3, SCC9, and SCC68, were injected superficially into the tongue of nude mice, allowed to grow for 1 month, and resected. Gross photography of HSC3 tumors in the tongue are shown in Supplemental Figure S2, J and K. Paraffin sections of tumor xenografts were stained with H&E (Supplemental Figure S2, A–F). NRP1 receptor protein, as visualized by IHC, was highly expressed in all three human OSCC lines analyzed in the orthotopic tongue environment (Supplemental Figure S2, G–I).

Blood vessels were detected in tissues by immunostaining for CD31 (Figure 5, A, C, E, F, and G). Blood vessel density, defined as the number of CD31+ vessels per square millimeter, was measured in each OSCC xenograft and in normal tongue (Figure 5, E–G). Quantification of blood vessel density did not differ significantly among the three OSCC lines examined (Figure 4E); however, the localization of NRP1 protein relative to blood vessel location was strikingly different between normal tongue and OSCC. In normal tongue, Nrp1 is found only in the outer epithelium, distant to underlying mucosal blood vessels (Figure 5, A and B). In contrast, blood vessels in squamous cell carcinomas appear to infiltrate the tumor in a pattern overlapping with areas of high NRP1 expression in tumor cells (Figure 5, C and D).

Figure 5.

Figure 5

Open in a new tab

NRP1 localization correlates with high vessel density in OSCC. Paraffin sections of mouse tongue (A and B) and human OSCC xenografts implanted in nude mouse tongue (C–G) were stained by immunohistochemistry for mouse CD31 (A, C, E–G), mouse Nrp1 (B), or human NRP1 (D). Brown color denotes immunostaining, and nuclei are stained blue with hematoxylin. In normal tongue, blood vessels are found in the mucosal and muscle layers, and Nrp1 is found in the outer epithelium (compare A and B). In contrast, in OSCC xenografts, blood vessels tend to infiltrate the tumor in areas with high NRP1 expression (compare serial sections C and D). Normal tongue has a high vascular density (A), whereas OSCC xenografts (E–G) have a lower vascular density than mouse tongue. Blood vessel density is quantified in Figure 4E. A yellow dotted line is drawn to denote the tumor/tongue border with tumor on the bottom in each panel. Scale bar = 100 μm. Nrp1, mouse neuropilin-1; NRP1, human neuropilin-1; OSCC, oral squamous cell carcinoma.

LVD, defined as the number of podoplanin+ vessels per square millimeter, was dramatically higher in HSC3 xenografts than in normal mouse tongue (Figure 6, A and B). Quantification of LVD was also significantly higher in HSC3-derived tumors than in either SCC9- or SCC68-derived tumors (Figure 4F and Figure 6, C–E).

Figure 6.

Figure 6

Open in a new tab

OSCC xenografts promote massive lymphangiogenesis in mouse tongue. Paraffin sections of nude mouse tongue (A) and human OSCC xenografts implanted in nude mouse tongue (B–E) were stained by immunohistochemistry for mouse Pdpn (brown). Sections were counterstained with hematoxylin (blue). Entire sagittal sections were photographed (A and B). Quantification of LVD is shown in Figure 4F. HSC3 has a higher LVD than SCC9 or SCC68. Scale bars: 1 mm (A and B); 100 μm (C–E). LVD, lymphatic vessel density; OSCC, oral squamous cell carcinoma; Pdpn, podoplanin.

Discussion

VEGF is the most widely studied angiogenic protein to date. In endothelial cells, VEGF function is mediated by two tyrosine kinase receptors (VEGFR1, VEGFR2) and the VEGF coreceptors, NRP1 and NRP2. However, the only VEGF receptor expressed by normal epithelial cells is NRP1.8, 9, 28 We previously reported that VEGF is produced by differentiated epithelial cells in the skin in a pattern that overlaps with NRP1.9 In this study we examined the expression patterns of Vegf and Nrp1 in normal adult mouse tongue. Because VEGF is secreted and heparin binding, merely locating the VEGF protein with antibodies via IHC does not indicate which cell(s) generates the VEGF ligand. Therefore, we used the VEGFLacZ transgenic mouse model,29 which express the β-galactosidase reporter gene (LacZ) and a nuclear localization signal inserted in the 3′ untranslated region of the VEGF gene, to identify cells producing VEGF by their blue nuclei after X-Gal staining. Our results confirm that in mouse tongue the differentiated, suprabasal epithelial cells constitutively generate VEGF under homeostatic conditions. With the use of antibodies that detect only the full-length Nrp1 receptor (and not soluble Nrp1), we show an overlapping pattern of Nrp1 and Vegf in mouse and human tongue epithelium. Therefore, normal adult skin and tongue epithelium have similar patterns of VEGF and NRP1 receptor expression.

Because no active angiogenesis occurs under these quiescent conditions, constitutive VEGF production is presumed to act as a survival factor, either for epithelial cells and/or for underlying endothelial cells. Consistent with this premise, mice lacking Nrp1 in keratinocytes showed more apoptosis after environmental insult, suggesting that Nrp1 provided a survival advantage.30 In addition, one side effect observed in patients who receive bevacizumab, a monoclonal antibody that targets VEGF protein, is geographic tongue, a benign condition characterized by erythematous, well-demarcated, patchy lesions.31

To date, the expression of NRP1 is correlated with advanced tumor stage and/or progression in many carcinomas, including bladder,32 breast,33 colon,13, 34 lung,35 ovarian,36 and prostate.37 Patients with NRP1+ gastric carcinomas had a poorer outcome in the AVAGAST randomized phase 3 trial that used chemotherapy and bevacizumab.38 However, mice lacking Nrp1 in epithelial cells did not develop skin tumors after carcinogen treatment, whereas 100% of control mice expressing Nrp1 developed tumors within 25 weeks.39 Taken together, these data suggest that the expression of NRP1 exacerbates carcinogenesis.

We sought to examine the expression of NRP1 during oral cancer progression, specifically in OSCC of the tongue. Our results found a distinct and abrupt up-regulation of NRP1 receptor in basal and spinous epithelial cells of oral epithelial dysplasia. The sharp increase in NRP1 was observed in biopsies of patients with oral dysplastic epithelial tissue and in mouse carcinogen-induced oral dysplastic epithelial samples. The elevated NRP1 expression persisted throughout later stages of progression and was present in squamous cell carcinoma of the tongue.

Interestingly, the pattern of NRP1 localization in tumor cells within the OSCC xenografts implanted in mouse tongue coincided with the pattern of blood vessel infiltration into the tumors. The spatial association between NRP1 expression by the carcinoma cells and neighboring vessels is highly reminiscent of that between NRP1-transfected tumors, which were shown to sequester VEGF in the tumor microenvironment, and neovessels.13 We describe this phenomenon as the depot effect or reservoir effect of NRP1 and propose that tumor-derived NRP1 protein holds VEGF protein (and other proangiogenic NRP1-associated ligands such as PlGF, HGF) on the surface of the tumor cell similar to heparan sulfate proteoglycans, creating a gradient of angiogenic molecules that function to attract blood vessel infiltration.

The pattern of NRP1 expression in CSCC dramatically differed from that observed in OSCC. CSCC, a common form of skin cancer that is lowly invasive and rarely metastatic,40 expresses NRP1 only in differentiated cells of the tumor.9 In CSCC derived from K14-HPV16 transgenic mice and in human patient tissue microarray samples, NRP1 expression did not correlate with disease progression or stage but was associated with the degree of differentiation.9 Of note, NRP1 was neither up-regulated in cutaneous dysplastic epithelium nor did its expression correlate with microvessel density in CSCC.9

However, OSCC, the most common cancer diagnosed in the oral cavity, is locally invasive, metastasizes early to lymph nodes, and recurs soon after treatment.41 Blood vessel density and LVD correlate with lymph node metastasis in human OSCC, and survival rates are lower in patients with high LVD.4 Expression of VEGF-A and VEGF-C correlates with lymph node metastasis in OSCC,42 and it is reported that NRP1 correlates with lymph node metastasis and poor prognosis in OSCC patients43 and is up-regulated in SCC of the esophagus.44 Herein, we demonstrate that OSCC cell lines and oral dysplastic patient biopsies strongly express NRP1 and that the localization of the NRP1 receptor coincides with VEGF expression and correlates with a highly invasive and motile potential in OSCC cell lines. Both VEGF expression and NRP1 expression were highest in the HSC3 tumor line, and LVD was highest in the HSC3 tumor xenografts, compared with other OSCC tumors. In vivo, areas of tumor NRP1 expression overlapped with blood vessel infiltration and correlated with higher lymphatic vessel density.

Another striking difference between OSCC and CSCC is that oral tumors strongly express NRP1 in their tumor-associated vessels,45 whereas skin tumors do not.9 Recently, miR-320 was found to target and silence NRP1 expression.45 Treatment of human OSCC ectopic xenografts with lentivirus that contains miR-320 precursor decreased tumor volume and vascular density.45 Alternately, transfection of the NRP1 construct into human OSCC cells induced an epithelial-to-mesenchymal transition in the tumor cells and increased their basal motility and invasiveness.43

The function of NRP1 in tumor cells is controversial and may depend on the expression of other receptors and ligands within a given type of tumor cell. In general, carcinoma cells (and the epithelial cells from which they originate) lack the tyrosine kinase VEGFR28, 9, 28; therefore, direct autocrine signaling through VEGF and the coreceptor complex of NRP1 and VEGFR2 cannot occur. However, exceptions do exist, and some highly metastatic carcinomas up-regulate VEGFR2 in addition to NRP1.46 Moreover, some cancer stem cells express both NRP1 and VEGFR2, and NRP1 is essential to maintain their stemness.39 Alternately, an autocrine loop between VEGF and VEGFR1/NRP1 may exist in some tumor cells such as those induced by HPV8.28 A recent study found that up-regulation of VEGFR isoforms (R1, R2, or R3) may be more prevalent in OSCC tumor cells than in other cancers with 88% of OSCC samples examined being positive for at least one of the isoforms.47 However, in a study in which OSCC xenografts were treated with a soluble VEGFR delivered via adenovirus, the researchers concluded that overall tumor burden was decreased solely by inhibiting angiogenesis and not via direct antitumor effects.48

In addition, NRP1 may be able to signal in the absence of canonical VEGFR signal transduction. A recent study reported that autocrine VEGF promotes tumor cell proliferation by signaling directly through the cytoplasmic domain of the NRP1 receptor, which interacts with GIPC1/synectin and the RhoGEF/Syx to activate RhoA and degrade p27.49 NRP1 was also found to signal independently of VEGFR1 in response to PlGF in medulloblastoma cells.50 However, other researchers found that the cytoplasmic domain of NRP1 was dispensable for its tumor-promoting activity in cancer cells, and the NRP1 extracellular domain was sufficient to regulate EGFR trafficking and downstream signaling through Akt to promote cell viability.51 Overall, these studies suggest that the interaction between NRP1 and synectin is key to its signaling potential and may be important for some coreceptor kinases but not others.

In the present study, we did not detect VEGFR2 in OSCC tumor cells (data not shown) and our data do not support an autocrine role for VEGF in the proliferation of OSCC cells because treatment with SEMA3A, a direct competitive inhibitor of VEGF binding to NRP1, did not affect tumor cell proliferation. However, SEMA3A did inhibit human OSCC cell migration similar to its effect seen on normal epithelial cells,8 suggesting that the NRP1 receptor in OSCC cells is bioactive. The inhibition of migration by SEMA3A is likely mediated through its binding to the NRP1 receptor and the subsequent interaction of the SEMA3A/NRP1 complex with plexin A1 as previously described.8, 52 Overall, our data suggest a protumorigenic role for NRP1 that is mediated by sequestering proangiogenic ligands (VEGF, PlGF, or HGF) in the tumor microenvironment and delivery of these ligands in a paracrine or juxtacrine fashion to neighboring vascular and lymphatic endothelial cells.

Interestingly, SEMA3A expression was found to be down-regulated in tongue cancer compared with normal tongue, and longer survival times were associated with higher SEMA3A expression levels.53 Overexpression of SEMA3A in tumor cells was shown to inhibit tumor angiogenesis and tumor cell migration and metastasis.54 Some antiangiogenic agents, such as sunitinib, were shown to cause massive tumor hypoxia that shifts tumor cells toward an epithelial-to-mesenchymal transition.54 When SEMA3A was overexpressed in cancer cells during sunitinib treatment, the deleterious effects on dissemination were counteracted.54 Although we did not directly compare NRP1 levels in CSCC and OSCC cell lines, we predict from our IHC results that OSCC expresses higher levels of NRP1 and would therefore be more responsive to inhibition by SEMA3A therapy in vivo. Exogenous SEMA3A protein drug therapy has yet to be shown in preclinical trials but may represent a promising addition to current therapies for highly aggressive OSCC.

Acknowledgments

We thank Ricardo Sanchez for histology sections and H&E staining, Kristin Johnson for photographic services, and Dr. James Rheinwald (Harvard Medical School, Brigham & Women's Hospital), Dr. Karl Munger (Harvard Medical School, Brigham & Women's Hospital), and Dr. Michael Klagsbrun (Harvard Medical School, Boston Children's Hospital) for cell lines. D.R.B. is the guarantor of this work and, as such, had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Footnotes

Supported by the Vascular Biology Program at Boston Children's Hospital and NIH National Cancer Institute and National Eye Institute grants K01CA118732 and R21CA155728 (D.R.B.), R01CA148633 (D.P.), and R01EY015435 (P.A.D.). Generation and production of the murine oral cancer tissue microarray was supported in part by the L-Core, Center for Oral Health Research, Medical University of South Carolina, funded by NIH National Institute of General Medical Sciences grant P30GM103331.

The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Disclosures: None declared.

Supplemental material for this article can be found at http://dx.doi.org/10.1016/j.ajpath.2015.11.021.

Supplemental Data

Supplemental Figure S1

Histology of carcinogen-induced mouse lingual tumors in a TMA. Mouse tumor TMA were purchased from the COHR, Medical University of South Carolina. H&E images were provided from the COHR. H&E images in each panel correspond to images in Figure 2: A: Normal mouse tongue. B and C: Epithelial dysplasia in the mouse tongue. D and E: Oral squamous cell carcinoma in the mouse tongue. F: Scan of entire TMA. Image was captured using a slide scanner. Original magnification, ×40 (A–F). COHR, Center for Oral Health Research; H&E, hematoxylin and eosin; TMA, tissue microarray.

mmc1.pdf (1.2MB, pdf)
Supplemental Figure S2

NRP1 expression in OSCC xenografts. A–I: Paraffin sections of human OSCC xenografts implanted in nude mouse tongue were stained by H&E (A–F) or immunohistochemistry with the use of anti-human NRP1 antibody (brown color) (G–I). Entire sagittal sections were photographed (A–C). All tumors express NRP1 in vivo (G–I). J and K: Gross image of mouse tongue with HSC3 tumor, dorsal view (J) and ventral view (K); dotted white line outlines the tumor margin. L: Western blot analysis of whole cell lysates from NOK, HSC3, and SCC9 are compared for NRP1 expression. NRP1 expression is up-regulated in oral cancer lines compared with NOK. Scale bars: 100 μm (D–I); 2 mm (J and K). Original magnification: ×20 (A–C). H&E, hematoxylin and eosin; NOK, normal oral keratinocyte; NRP1, neuropilin 1; OSCC, oral squamous cell carcinoma.

mmc2.pdf (17.5MB, pdf)
Supplemental Table S1
mmc3.docx (17.3KB, docx)

References

  • 1.Howlader N.N.A., Krapcho M., Garshell J., Neyman N., Altekruse S.F., Kosary C.L., Yu M., Ruhl J., Tatalovich Z., Cho H., Mariotto A., Lewis D.R., Chen H.S., Feuer E.J., Cronin K.A., editors. SEER Cancer Statistics Review, 1975-2010. National Cancer Institute; Bethesda, MD: 2013. http://seer.cancer.gov/archive/csr/1975_2010 Available at. (accessed November 11, 2015) [Google Scholar]
  • 2.Karia P.S., Han J., Schmults C.D. Cutaneous squamous cell carcinoma: estimated incidence of disease, nodal metastasis, and deaths from disease in the United States, 2012. J Am Acad Dermatol. 2013;68:957–966. doi: 10.1016/j.jaad.2012.11.037. [DOI] [PubMed] [Google Scholar]
  • 3.Folkman J. Is angiogenesis an organizing principle in biology and medicine? J Pediatr Surg. 2007;42:1–11. doi: 10.1016/j.jpedsurg.2006.09.048. [DOI] [PubMed] [Google Scholar]
  • 4.Miyahara M., Tanuma J., Sugihara K., Semba I. Tumor lymphangiogenesis correlates with lymph node metastasis and clinicopathologic parameters in oral squamous cell carcinoma. Cancer. 2007;110:1287–1294. doi: 10.1002/cncr.22900. [DOI] [PubMed] [Google Scholar]
  • 5.Bielenberg D., Kurschat P., Klagsbrun M. In: Neuropilins, receptors central to angiogenesis and neuronal guidance. Aird W.C., editor. Cambridge University Press; Cambridge, MA: 2007. pp. 327–334. [Google Scholar]
  • 6.Raimondi C., Ruhrberg C. Neuropilin signalling in vessels, neurons and tumours. Semin Cell Dev Biol. 2013;24:172–178. doi: 10.1016/j.semcdb.2013.01.001. [DOI] [PubMed] [Google Scholar]
  • 7.Gagnon M.L., Bielenberg D.R., Gechtman Z., Miao H.Q., Takashima S., Soker S., Klagsbrun M. Identification of a natural soluble neuropilin-1 that binds vascular endothelial growth factor: in vivo expression and antitumor activity. Proc Natl Acad Sci U S A. 2000;97:2573–2578. doi: 10.1073/pnas.040337597. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Kurschat P., Bielenberg D., Rossignol-Tallandier M., Stahl A., Klagsbrun M. Neuron restrictive silencer factor NRSF/REST is a transcriptional repressor of neuropilin-1 and diminishes the ability of semaphorin 3A to inhibit keratinocyte migration. J Biol Chem. 2006;281:2721–2729. doi: 10.1074/jbc.M507860200. [DOI] [PubMed] [Google Scholar]
  • 9.Shahrabi-Farahani S., Wang L., Zwaans B.M., Santana J.M., Shimizu A., Takashima S., Kreuter M., Coultas L., D'Amore P.A., Arbeit J.M., Akslen L.A., Bielenberg D.R. Neuropilin 1 expression correlates with differentiation status of epidermal cells and cutaneous squamous cell carcinomas. Lab Invest. 2014;94:752–765. doi: 10.1038/labinvest.2014.66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Bielenberg D.R., Pettaway C.A., Takashima S., Klagsbrun M. Neuropilins in neoplasms: expression, regulation, and function. Exp Cell Res. 2006;312:584–593. doi: 10.1016/j.yexcr.2005.11.024. [DOI] [PubMed] [Google Scholar]
  • 11.Wild J.R., Staton C.A., Chapple K., Corfe B.M. Neuropilins: expression and roles in the epithelium. Int J Exp Pathol. 2012;93:81–103. doi: 10.1111/j.1365-2613.2012.00810.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Miao H.Q., Lee P., Lin H., Soker S., Klagsbrun M. Neuropilin-1 expression by tumor cells promotes tumor angiogenesis and progression. FASEB J. 2000;14:2532–2539. doi: 10.1096/fj.00-0250com. [DOI] [PubMed] [Google Scholar]
  • 13.Parikh A.A., Fan F., Liu W.B., Ahmad S.A., Stoeltzing O., Reinmuth N., Bielenberg D., Bucana C.D., Klagsbrun M., Ellis L.M. Neuropilin-1 in human colon cancer: expression, regulation, and role in induction of angiogenesis. Am J Pathol. 2004;164:2139–2151. doi: 10.1016/S0002-9440(10)63772-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Miao H.Q., Klagsbrun M. Neuropilin is a mediator of angiogenesis. Cancer Metastasis Rev. 2000;19:29–37. doi: 10.1023/a:1026579711033. [DOI] [PubMed] [Google Scholar]
  • 15.Bielenberg D.R., Klagsbrun M. Targeting endothelial and tumor cells with semaphorins. Cancer Metastasis Rev. 2007;26:421–431. doi: 10.1007/s10555-007-9097-4. [DOI] [PubMed] [Google Scholar]
  • 16.Otsubo T., Hida Y., Ohga N., Sato H., Kai T., Matsuki Y., Takasu H., Akiyama K., Maishi N., Kawamoto T., Shinohara N., Nonomura K., Hida K. Identification of novel targets for antiangiogenic therapy by comparing the gene expressions of tumor and normal endothelial cells. Cancer Sci. 2014;105:560–567. doi: 10.1111/cas.12394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Matsui T., Ota T., Ueda Y., Tanino M., Odashima S. Isolation of a highly metastatic cell line to lymph node in human oral squamous cell carcinoma by orthotopic implantation in nude mice. Oral Oncol. 1998;34:253–256. [PubMed] [Google Scholar]
  • 18.Rheinwald J.G., Beckett M.A. Tumorigenic keratinocyte lines requiring anchorage ad fibroblast support cultured from human squamous cell carcinomas. Cancer Res. 1981;41:1657–1663. [PubMed] [Google Scholar]
  • 19.Lin C.J., Grandis J.R., Carey T.E., Gollin S.M., Whiteside T.L., Koch W.M., Ferris R.L., Lai S.Y. Head and neck squamous cell carcinoma cell lines: established models and rationale for selection. Head Neck. 2007;29:163–188. doi: 10.1002/hed.20478. [DOI] [PubMed] [Google Scholar]
  • 20.Degen M., Natarajan E., Barron P., Widlund H.R., Rheinwald J.G. MAPK/ERK-dependent translation factor hyperactivation and dysregulated laminin gamma2 expression in oral dysplasia and squamous cell carcinoma. Am J Pathol. 2012;180:2462–2478. doi: 10.1016/j.ajpath.2012.02.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Piboonniyom S.O., Timmermann S., Hinds P., Munger K. Aberrations in the MTS1 tumor suppressor locus in oral squamous cell carcinoma lines preferentially affect the INK4A gene and result in increased cdk6 activity. Oral Oncol. 2002;38:179–186. doi: 10.1016/s1368-8375(01)00042-2. [DOI] [PubMed] [Google Scholar]
  • 22.Soker S., Takashima S., Miao H.Q., Neufeld G., Klagsbrun M. Neuropilin-1 is expressed by endothelial and tumor cells as an isoform-specific receptor for vascular endothelial growth factor. Cell. 1998;92:735–745. doi: 10.1016/s0092-8674(00)81402-6. [DOI] [PubMed] [Google Scholar]
  • 23.Bielenberg D.R., Shimizu A., Klagsbrun M. Semaphorin-induced cytoskeletal collapse and repulsion of endothelial cells. Methods Enzymol. 2008;443:299–314. doi: 10.1016/S0076-6879(08)02015-6. [DOI] [PubMed] [Google Scholar]
  • 24.Berge M., Allanic D., Bonnin P., de Montrion C., Richard J., Suc M., Boivin J.F., Contreres J.O., Lockhart B.P., Pocard M., Levy B.I., Tucker G.C., Tobelem G., Merkulova-Rainon T. Neuropilin-1 is upregulated in hepatocellular carcinoma and contributes to tumour growth and vascular remodelling. J Hepatol. 2011;55:866–875. doi: 10.1016/j.jhep.2011.01.033. [DOI] [PubMed] [Google Scholar]
  • 25.Banyard J., Chung I., Migliozzi M., Phan D.T., Wilson A.M., Zetter B.R., Bielenberg D.R. Identification of genes regulating migration and invasion using a new model of metastatic prostate cancer. BMC Cancer. 2014;14:387. doi: 10.1186/1471-2407-14-387. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Miao H.Q., Soker S., Feiner L., Alonso J.L., Raper J.A., Klagsbrun M. Neuropilin-1 mediates collapsin-1/semaphorin III inhibition of endothelial cell motility: functional competition of collapsin-1 and vascular endothelial growth factor-165. J Cell Biol. 1999;146:233–242. doi: 10.1083/jcb.146.1.233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Guttmann-Raviv N., Shraga-Heled N., Varshavsky A., Guimaraes-Sternberg C., Kessler O., Neufeld G. Semaphorin-3A and semaphorin-3F work together to repel endothelial cells and to inhibit their survival by induction of apoptosis. J Biol Chem. 2007;282:26294–26305. doi: 10.1074/jbc.M609711200. [DOI] [PubMed] [Google Scholar]
  • 28.Ding X., Lucas T., Marcuzzi G.P., Pfister H., Eming S.A. Distinct functions of epidermal and myeloid-derived VEGF-A in skin tumorigenesis mediated by HPV8. Cancer Res. 2015;75:330–343. doi: 10.1158/0008-5472.CAN-13-3007. [DOI] [PubMed] [Google Scholar]
  • 29.Maharaj A.S., Saint-Geniez M., Maldonado A.E., D'Amore P.A. Vascular endothelial growth factor localization in the adult. Am J Pathol. 2006;168:639–648. doi: 10.2353/ajpath.2006.050834. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Riese A., Eilert Y., Meyer Y., Arin M., Baron J.M., Eming S., Krieg T., Kurschat P. Epidermal expression of neuropilin 1 protects murine keratinocytes from UVB-induced apoptosis. PLoS One. 2012;7:e50944. doi: 10.1371/journal.pone.0050944. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Gavrilovic I.T., Balagula Y., Rosen A.C., Ramaswamy V., Dickler M.N., Dunkel I.J., Lacouture M.E. Characteristics of oral mucosal events related to bevacizumab treatment. Oncologist. 2012;17:274–278. doi: 10.1634/theoncologist.2011-0198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Cheng W., Fu D., Wei Z.F., Xu F., Xu X.F., Liu Y.H., Ge J.P., Tian F., Han C.H., Zhang Z.Y., Zhou L.M. NRP-1 expression in bladder cancer and its implications for tumor progression. Tumour Biol. 2014;35:6089–6094. doi: 10.1007/s13277-014-1806-3. [DOI] [PubMed] [Google Scholar]
  • 33.Jubb A.M., Strickland L.A., Liu S.D., Mak J., Schmidt M., Koeppen H. Neuropilin-1 expression in cancer and development. J Pathol. 2012;226:50–60. doi: 10.1002/path.2989. [DOI] [PubMed] [Google Scholar]
  • 34.Hansel D.E., Wilentz R.E., Yeo C.J., Schulick R.D., Montgomery E., Maitra A. Expression of neuropilin-1 in high-grade dysplasia, invasive cancer, and metastases of the human gastrointestinal tract. Am J Surg Pathol. 2004;28:347–356. doi: 10.1097/00000478-200403000-00007. [DOI] [PubMed] [Google Scholar]
  • 35.Hong T.M., Chen Y.L., Wu Y.Y., Yuan A., Chao Y.C., Chung Y.C., Wu M.H., Yang S.C., Pan S.H., Shih J.Y., Chan W.K., Yang P.C. Targeting neuropilin 1 as an antitumor strategy in lung cancer. Clin Cancer Res. 2007;13:4759–4768. doi: 10.1158/1078-0432.CCR-07-0001. [DOI] [PubMed] [Google Scholar]
  • 36.Osada R., Horiuchi A., Kikuchi N., Ohira S., Ota M., Katsuyama Y., Konishi I. Expression of semaphorins, vascular endothelial growth factor, and their common receptor neuropilins and alleic loss of semaphorin locus in epithelial ovarian neoplasms: increased ratio of vascular endothelial growth factor to semaphorin is a poor prognostic factor in ovarian carcinomas. Hum Pathol. 2006;37:1414–1425. doi: 10.1016/j.humpath.2006.04.031. [DOI] [PubMed] [Google Scholar]
  • 37.Li Q., Li Q., Nuccio J., Liu C., Duan P., Wang R., Jones L.W., Chung L.W., Zhau H.E. Metastasis initiating cells in primary prostate cancer tissues from transurethral resection of the prostate (TURP) predicts castration-resistant progression and survival of prostate cancer patients. Prostate. 2015;75:1312–1321. doi: 10.1002/pros.23011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Van Cutsem E., de Haas S., Kang Y.K., Ohtsu A., Tebbutt N.C., Ming Xu J., Peng Yong W., Langer B., Delmar P., Scherer S.J., Shah M.A. Bevacizumab in combination with chemotherapy as first-line therapy in advanced gastric cancer: a biomarker evaluation from the AVAGAST randomized phase III trial. J Clin Oncol. 2012;30:2119–2127. doi: 10.1200/JCO.2011.39.9824. [DOI] [PubMed] [Google Scholar]
  • 39.Beck B., Driessens G., Goossens S., Youssef K.K., Kuchnio A., Caauwe A., Sotiropoulou P.A., Loges S., Lapouge G., Candi A., Mascre G., Drogat B., Dekoninck S., Haigh J.J., Carmeliet P., Blanpain C. A vascular niche and a VEGF-Nrp1 loop regulate the initiation and stemness of skin tumours. Nature. 2011;478:399–403. doi: 10.1038/nature10525. [DOI] [PubMed] [Google Scholar]
  • 40.Rogers H.W., Weinstock M.A., Harris A.R., Hinckley M.R., Feldman S.R., Fleischer A.B., Coldiron B.M. Incidence estimate of nonmelanoma skin cancer in the United States, 2006. Arch Dermatol. 2010;146:283–287. doi: 10.1001/archdermatol.2010.19. [DOI] [PubMed] [Google Scholar]
  • 41.Bello I.O., Soini Y., Salo T. Prognostic evaluation of oral tongue cancer: means, markers and perspectives (I) Oral Oncol. 2010;46:630–635. doi: 10.1016/j.oraloncology.2010.06.006. [DOI] [PubMed] [Google Scholar]
  • 42.Yanase M., Kato K., Yoshizawa K., Noguchi N., Kitahara H., Nakamura H. Prognostic value of vascular endothelial growth factors A and C in oral squamous cell carcinoma. J Oral Pathol Med. 2014;43:514–520. doi: 10.1111/jop.12167. [DOI] [PubMed] [Google Scholar]
  • 43.Chu W., Song X., Yang X., Ma L., Zhu J., He M., Wang Z., Wu Y. Neuropilin-1 promotes epithelial-to-mesenchymal transition by stimulating nuclear factor-kappa B and is associated with poor prognosis in human oral squamous cell carcinoma. PLoS One. 2014;9:e101931. doi: 10.1371/journal.pone.0101931. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Alattar M., Omo A., Elsharawy M., Li J. Neuropilin-1 expression in squamous cell carcinoma of the oesophagus. Eur J Cardiothorac Surg. 2014;45:514–520. doi: 10.1093/ejcts/ezt380. [DOI] [PubMed] [Google Scholar]
  • 45.Wu Y.Y., Chen Y.L., Jao Y.C., Hsieh I.S., Chang K.C., Hong T.M. miR-320 regulates tumor angiogenesis driven by vascular endothelial cells in oral cancer by silencing neuropilin 1. Angiogenesis. 2014;17:247–260. doi: 10.1007/s10456-013-9394-1. [DOI] [PubMed] [Google Scholar]
  • 46.Migliozzi M.H.Y., Seth M., Brown G., Kwan J., Coma S., Panigrahy D., Adam R.M., Banyard J., Shimizu A., Bielenberg D.R. VEGF/VEGFR2 autocrine signaling stimulates metastasis in prostate cancer cells. Curr Angiogenesis. 2014;3:231–244. [Google Scholar]
  • 47.Pianka A., Knosel T., Probst F.A., Troeltzsch M., Woodlock T., Otto S., Ehrenfeld M., Troeltzsch M. Vascular endothelial growth factor receptor isoforms: are they present in oral squamous cell carcinoma? J Oral Maxillofac Surg. 2015;73:897–904. doi: 10.1016/j.joms.2014.12.030. [DOI] [PubMed] [Google Scholar]
  • 48.Okada Y., Ueno H., Katagiri M., Oneyama T., Shimomura K., Sakurai S., Mataga I., Moride M., Hasegawa H. Experimental study of antiangiogenic gene therapy targeting VEGF in oral cancer. Odontology. 2010;98:52–59. doi: 10.1007/s10266-009-0117-4. [DOI] [PubMed] [Google Scholar]
  • 49.Yoshida A., Shimizu A., Asano H., Kadonosono T., Kondoh S.K., Geretti E., Mammoto A., Klagsbrun M., Seo M.K. VEGF-A/NRP1 stimulates GIPC1 and Syx complex formation to promote RhoA activation and proliferation in skin cancer cells. Biol Open. 2015;4:1063–1076. doi: 10.1242/bio.010918. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Snuderl M., Batista A., Kirkpatrick N.D., Ruiz de Almodovar C., Riedemann L., Walsh E.C., Anolik R., Huang Y., Martin J.D., Kamoun W., Knevels E., Schmidt T., Farrar C.T., Vakoc B.J., Mohan N., Chung E., Roberge S., Peterson T., Bais C., Zhelyazkova B.H., Yip S., Hasselblatt M., Rossig C., Niemeyer E., Ferrara N., Klagsbrun M., Duda D.G., Fukumura D., Xu L., Carmeliet P., Jain R.K. Targeting placental growth factor/neuropilin 1 pathway inhibits growth and spread of medulloblastoma. Cell. 2013;152:1065–1076. doi: 10.1016/j.cell.2013.01.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Rizzolio S., Rabinowicz N., Rainero E., Lanzetti L., Serini G., Norman J., Neufeld G., Tamagnone L. Neuropilin-1-dependent regulation of EGF-receptor signaling. Cancer Res. 2012;72:5801–5811. doi: 10.1158/0008-5472.CAN-12-0995. [DOI] [PubMed] [Google Scholar]
  • 52.Shimizu A., Mammoto A., Italiano J.E., Jr., Pravda E., Dudley A.C., Ingber D.E., Klagsbrun M. ABL2/ARG tyrosine kinase mediates SEMA3F-induced RhoA inactivation and cytoskeleton collapse in human glioma cells. J Biol Chem. 2008;283:27230–27238. doi: 10.1074/jbc.M804520200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Song X., Zhang W., Zhang Y., Zhang H., Fu Z., Ye J., Liu L., Song X., Wu Y. Expression of semaphorin 3A and neuropilin 1 with clinicopathological features and survival in human tongue cancer. Med Oral Patol Oral Cir Bucal. 2012;17:e962–e968. doi: 10.4317/medoral.18168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Maione F., Capano S., Regano D., Zentilin L., Giacca M., Casanovas O., Bussolino F., Serini G., Giraudo E. Semaphorin 3A overcomes cancer hypoxia and metastatic dissemination induced by antiangiogenic treatment in mice. J Clin Invest. 2012;122:1832–1848. doi: 10.1172/JCI58976. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Figure S1

Histology of carcinogen-induced mouse lingual tumors in a TMA. Mouse tumor TMA were purchased from the COHR, Medical University of South Carolina. H&E images were provided from the COHR. H&E images in each panel correspond to images in Figure 2: A: Normal mouse tongue. B and C: Epithelial dysplasia in the mouse tongue. D and E: Oral squamous cell carcinoma in the mouse tongue. F: Scan of entire TMA. Image was captured using a slide scanner. Original magnification, ×40 (A–F). COHR, Center for Oral Health Research; H&E, hematoxylin and eosin; TMA, tissue microarray.

mmc1.pdf (1.2MB, pdf)
Supplemental Figure S2

NRP1 expression in OSCC xenografts. A–I: Paraffin sections of human OSCC xenografts implanted in nude mouse tongue were stained by H&E (A–F) or immunohistochemistry with the use of anti-human NRP1 antibody (brown color) (G–I). Entire sagittal sections were photographed (A–C). All tumors express NRP1 in vivo (G–I). J and K: Gross image of mouse tongue with HSC3 tumor, dorsal view (J) and ventral view (K); dotted white line outlines the tumor margin. L: Western blot analysis of whole cell lysates from NOK, HSC3, and SCC9 are compared for NRP1 expression. NRP1 expression is up-regulated in oral cancer lines compared with NOK. Scale bars: 100 μm (D–I); 2 mm (J and K). Original magnification: ×20 (A–C). H&E, hematoxylin and eosin; NOK, normal oral keratinocyte; NRP1, neuropilin 1; OSCC, oral squamous cell carcinoma.

mmc2.pdf (17.5MB, pdf)
Supplemental Table S1
mmc3.docx (17.3KB, docx)

Articles from The American Journal of Pathology are provided here courtesy of American Society for Investigative Pathology

RESOURCES