Neuropilin-1 in Human Colon CancerExpression, Regulation, and Role in Induction of Angiogenesis

Neuropilin-1 in Human Colon CancerExpression, Regulation, and Role in Induction of Angiogenesis
of 13
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  Neuropilin-1 in Human Colon Cancer Expression, Regulation, and Role in Induction of Angiogenesis Alexander A. Parikh,* Fan Fan, † Wen Biao Liu, † Syed A. Ahmad,* Oliver Stoeltzing, † Niels Reinmuth, † Diane Bielenberg, ‡ Corazon D. Bucana, † Michael Klagsbrun, ‡ andLee M. Ellis* † From the Departments of Surgical Oncology  *  and Cancer Biology, † The University of Texas M. D. Anderson Cancer Center,Houston, Texas; and the Department of Surgical Research, ‡ Children’s Hospital and Harvard Medical School,Boston, Massachusetts  Neuropilin-1 (NRP-1), a recently identified co-recep-tor for vascular endothelial growth factor, is ex-pressed by several nongastrointestinal tumor typesand enhances prostate cancer angiogenesis and growth in preclinical models. We investigated the ex-pression and regulation of NRP-1 and the effect of NRP-1 overexpression on angiogenesis and growth of human colon adenocarcinoma by immunohisto-chemistry and   in situ  hybridization. NRP-1 was ex-pressed in 20 of 20 human colon adenocarcinoma specimens but not in the adjacent nonmalignant co-lonic mucosa. By reverse transcriptase-polymerasechain reaction analysis, NRP-1 mRNA was expressed in seven of seven colon adenocarcinoma cell lines.Subcutaneous xenografts of stably transfected KM12SM/LM2 human colon cancer cells overexpress-ing NRP-1 led to increased tumor growth and angio-genesis in nude mice. In   in vitro  assays, conditioned medium from NRP-1-transfected cell lines led to an increase in endothelial cell migration, but did not affect endothelial cell growth. Epidermal growth fac-tor (EGF) led to induction of NRP-1 in human colon adenocarcinoma cells and selective blockade of theepidermal growth factor receptor (EGFR) decreased constitutive and EGF-induced NRP-1 expression.Blockade of the Erk 1/2 and P38 mitogen-activated protein kinase signaling pathways also led to a de-crease in constitutive and EGF-induced NRP-1 expres-sion. These findings demonstrate the ubiquitous ex-pression of NRP-1 in human colon cancer and suggest that NRP-1 may contribute to colon cancer angiogen-esis and growth. This study also suggests that EGF and mitogen-activated protein kinase signaling pathwaysplay an important role in NRP-1 regulation in colon cancer cells.  (Am J Pathol 2004, 164:2139–2151) The growth of cancers and the development of metasta-sis is angiogenesis-dependent. Of the many proangio-genic factors identified, vascular endothelial growth fac-tor (VEGF; also known as vascular permeability factor) isthe best characterized. VEGF has been associated withincreased angiogenesis and advanced-stage disease ina variety of solid tumor types including colon cancer. 1,2 The VEGF family of proteins are highly structurally relatedproteins including VEGF-A (commonly designated asVEGF), VEGF-B, VEGF-C, VEGF-D, and VEGF-E, andplacenta growth factor. 3–5 The most prominent and char-acterized member, VEGF-A, exists as different isoformsbased on the number of amino acids: VEGF 121 , VEGF 145 ,VEGF 165 , VEGF 189 , and VEGF 206 . Most studies suggestthat VEGF 165  is the most abundant and biologically activeisoform. 6,7 Members of the VEGF family act primarily viathree membrane-bound tyrosine kinase receptors: VEGFreceptor-1 (VEGFR-1; Flt-1), VEGFR-2 (Flk-1/KDR), andVEGFR-3 (Flt-4). 8 Although these receptors were initiallythought to be present only on endothelial cells, recentevidence suggests that VEGF receptors may also beinfrequently expressed on tumor cells. 9,10 Neuropilin (NRP)-1 was srcinally described as a 130-to 140-kd cell-surface glycoprotein expressed in the de-veloping  Xenopus laevis  nervous system. 11 Subsequently,it was discovered that this transmembrane glycoproteinis a receptor for the semaphorins/collapsins, a large fam-ily of secreted and transmembrane proteins that serve asguidance signals in axonal and neuronal develop-ment. 12–15 Several studies have also suggested a role forNRP-1 in embryological vasculogenesis and angiogene-sis. NRP-1 has been shown to be expressed in the de-veloping skeletal and cardiovascular systems in embry-os. 12,16 NRP-1 knockout mice suffer from insufficient anddelayed vascularization leading to embryonic death, 17,18 Supported, in part, by the National Institutes of Health (T-32 training grantCA-09599 to A.A.P. and S.A.A. and cancer center support grant CA-16672 to M.K.), the Gillson Longenbaugh Foundation (to L.M.E.), the Jonand Susie Hall Fund for Colon Cancer Research (to L.M.E.), and theNational Cancer Institute (grants 37392 and 45448 to M.K.).A.A.P. and F.F. contributed equally to this work.Accepted for publication February 12, 2004.Address reprint requests to Lee M. Ellis, M.D., Department of SurgicalOncology, Box 444, The University of Texas M. D. Anderson CancerCenter, 1515 Holcombe Blvd., Houston, TX 77030-4009. American Journal of Pathology, Vol. 164, No. 6, June 2004 Copyright © American Society for Investigative Pathology  2139  whereas overexpression of NRP-1 in transgenic mice islethal because of hemorrhage in the head and neck,excess blood vessel formation, and malformed hearts. 16 NRP-1 has also been found to be expressed in adultendothelial cells and, to a lesser degree, in a variety ofother tissues including lung, heart, liver, kidney, pan-creas, and placenta as well as in osteoblasts and bonemarrow stromal cells. 19,20 The specific functions of NRP-1 in vessel developmentand angiogenesis remain to be elucidated. 14,19,21 UnlikeVEGFR-1, VEGFR-2, and VEGFR-3, NRP-1 does not con-tain a tyrosine-kinase domain and therefore seems to actas a co-receptor for VEGF 165 . 19 The binding of VEGF 165 to NRP-1 is mediated by amino acids residing at thecarboxyl-terminal part of the exon 7-encoded peptide ofVEGF 165 . 19 In contrast, the binding of VEGF 165  toVEGFR-1 and VEGFR-2 occurs via exon 3 and exon 4,respectively, 19 thus enabling VEGF 165  to bind to bothNRP-1 and VEGFR-1 or VEGFR-2 simultaneously. Inhibi-tion of VEGF 165  binding to NRP-1 in endothelial cells alsodecreases VEGF 165  binding to VEGFR-2 and subsequentmitogenic activity. 22 Furthermore, co-transfection ofNRP-1 into VEGFR-2-expressing endothelial cells en-hances the binding of VEGF 165  to VEGFR-2 and subse-quentmitogenicandchemotacticactivityascomparedtocells expressing VEGFR-2 alone. 13,19 Endothelial cellsexpressing NRP-1 but not VEGFR-2 do not respond toany VEGF isoform, suggesting that NRP-1 is not a signal-ing receptor for chemotaxis, in and of itself, but ratheracts as a co-receptor for VEGFR-2, enhancing VEGF’sactivity as an angiogenic factor. 19 Expression of NRP-1 has recently been found in pros-tate cancer, breast cancer, and melanoma cell lines aswell as several tumor types from patient speci-mens. 12,23–25 Overexpression of NRP-1 in rat prostatecarcinoma cells results in increased tumor growth  in vivo as well as increased microvessel density and endothelialcell proliferation. 12,14 Prostate tumor cell NRP-1 expres-sion also enhances binding of VEGF 165  to these tumorcells. 12 Recent studies suggest that VEGF 165  has a directeffect on tumor cells mediated through NRP-1. VEGF 165 has been shown to act as an autocrine survival factor inNRP-1-positive breast carcinoma cells lacking VEGFR-2,likely occurring via activation of the PI-3 kinase path-way. 23 In studies on human tumor specimens, NRP-1expressioncorrelateswiththemetastaticpotential,stage,and grade of prostate cancer. NRP-1 has also beenfound to be associated with advanced stage and gradeof astrocytoma. 24,26 However, the expression and regulation of NRP-1 ingastrointestinal malignancies, including colorectal ade-nocarcinoma, has not been characterized. In this study,we investigated the expression of NRP-1 in human colonadenocarcinomas and uninvolved colonic mucosa, andthe effect of NRP-1 overexpression on tumor growth andangiogenesis. We also investigated mechanisms of in-duction of NRP-1 in human colon adenocarcinoma celllines. Materials and Methods  Tissue Specimens Specimens of colon adenocarcinoma and adjacent non-malignant colonic mucosa were obtained from 20 pa-tients at The University of Texas M. D. Anderson CancerCenter immediately after resection under a protocol ap-proved by the institutional review board at M. D. Ander-son Cancer Center. Specimens were either frozen inoptimum cutting temperature (OCT) solution (Miles,Elkhart, IN) and stored at  80°C or fixed in formalin andembedded in paraffin at the time of their collection. Thehistopathological diagnosis of colon adenocarcinomaand adjacent nonmalignant mucosa was confirmed bythe Department of Pathology. No patients received pre-operative chemotherapy or radiation therapy. Reagents and Chemicals Recombinant human epidermal growth factor (EGF), in-sulin-like growth factor-1 (IGF-1), interleukin-1   (IL-1  ),and tumor necrosis factor (TNF)-   were purchased fromR&D Systems (Minneapolis, MN). U0126, the extracellu-lar signal-regulated kinase 1/2 (Erk 1/2) mitogen-acti-vated protein kinase (MAPK) inhibitor, was obtained fromNew England Biolabs (Beverly, MA). The phosphatidyl-inositol-3 (PI-3) kinase inhibitor Wortmannin was pur-chased from Sigma Chemical Company (St. Louis, MO).The P38 MAPK inhibitor SB203580 was purchased fromCalbiochem (San Diego, CA). The anti-human EGF re-ceptor (EGFR) monoclonal antibody C225 was kindlyprovided by Daniel J. Hicklin, Ph.D. (ImClone Systems,New York, NY). Cell Lines The human colon adenocarcinoma cell lines HT29,SW480, SW620, and RKO, and human umbilical veinendothelial cells (HUVECs) were obtained from the Amer-ican Type Culture Collection (Manassas, VA). The humancolon adenocarcinoma cell lines KM12L4, KM12SM,KM12SMLM2, and KM20 were provided by IJ Fidler,D.V.M., Ph.D. (M. D. Anderson Cancer Center). Coloncancer cell lines were cultured and maintained in minimalessential medium (MEM) supplemented with 10% fetalbovine serum (FBS), 2 U/ml penicillin-streptomycin, vita-mins, 1 mmol/L sodium pyruvate, 2 mmol/L  L -glutamine,and nonessential amino acids at 37°C in 5% CO 2  and95% air. HUVECs were plated onto 0.5% gelatin-coatedflasks and maintained in MEM supplemented with 15%FBS, 2 U/ml penicillin-streptomycin, vitamins, 1 mmol/Lsodiumpyruvate,2mmol/L L -glutamine,andnonessentialamino acids at 37°C in 5% CO 2  and 95% air. Immunohistochemical Staining for NRP-1 inFrozen Tissue Specimens Tissue specimens frozen in OCT solution were sectioned(8 to 10   m thick), mounted on positively charged Super- 2140 Parikh et al AJP June 2004, Vol. 164, No. 6   frost slides (Fisher Scientific, Houston, TX), and air-driedfor 30 minutes. Sections were fixed in cold acetone (5minutes) followed by 1:1 acetone:chloroform (5 minutes)and then acetone (5 minutes) and washed with phos-phate-buffered saline (PBS) three times for 3 minuteseach time. All samples were incubated with 3% hydrogenperoxide in methanol for 12 minutes at room temperatureto block endogenous peroxidase. Sections were thenwashed three times for 3 minutes each time with PBS (pH7.5) and incubated for 20 minutes at room temperature ina protein-blocking solution consisting of PBS supple-mented with 1% normal goat serum and 5% normal horseserum. The primary antibody directed against NRP-1,polyclonal rabbit anti-NRP-1 (1:150 dilution; Santa CruzBiotechnology Inc., Santa Cruz, CA), was applied to thesections, which were incubated overnight at 4°C. Sec-tions were then rinsed three times for 3 minutes in PBSandincubatedfor10minutesinprotein-blockingsolution.The secondary antibody [peroxidase-conjugated goatanti-rabbit IgG (H  L), Jackson Research Laboratories,Westgrove, PA] was then used at a 1:400 dilution. Sec-tions were washed three times with PBS followed byrinsing with PBS/0.1% Brij detergent. This was followedby incubation with stable diaminobenzidine substrate(Research Genetics, Huntsville, AL), during which thestaining was monitored by bright-field microscopy. Thereaction was halted by rinsing with ddH 2 O. The sectionswere counterstained with Gill’s no. 3 hematoxylin solution(Sigma Chemical), mounted with Universal Mount (Re-search Genetics), and analyzed with light microscopy.The protocol for control specimens was similar exceptthat the primary antibody was omitted. Immunofluorescent Staining for NRP-1 and CK-22 in Frozen Tissue Specimens Tissue specimens frozen in OCT were sectioned andprocessed as described above. Immunofluorescentstaining for NRP-1, cytokeratin-22 (CK-22, an epithelialcell marker), and NRP-1/CK-22 double staining were per-formed in the manner previously described. 27 In Situ  Hybridization In situ  hybridization was performed as previously de-scribed. 28 An anti-sense NRP-1-specific riboprobe wasmade from a 750-bp 3  UTR cDNA fragment in BluescriptII KS(  ) (Stratagene, La Jolla, CA) by using a digoxigeninRNA-labeling kit (Boehringer Mannheim, Indianapolis, IN)and used in  in situ  hybridization as previously de-scribed. 29 RNA Extraction and Northern Blot Analysis Total RNA was harvested from subconfluent tumor cellsin culture by using Trizol Reagent (Life Technologies,Inc., Grand Island, NY) following the manufacturer’s in-structions. Northern blot analysis was performed as pre-viously described. 30 Briefly, 25   g of RNA was fraction-ated on 1% denaturing formaldehyde/agarose gels andtransferred to a Hybond-N   positively charged nylonmembrane (Amersham Biosciences, Piscataway, NJ)overnightbycapillaryelution.Afterultravioletcrosslinkingat 120,000 mJ/cm 2 with an ultraviolet Stratalinker 1800(Stratagene),themembraneswereprehybridizedat65°Cfor 3 to 4 hours in rapid hybridization buffer (Amersham).The membranes were then hybridized at 65°C overnightwith the cDNA probe for NRP-1 or glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The probed mem-branes were then washed and autoradiography per-formed. The cDNA probes used comprised a humanNRP-1 450-bp cDNA probe derived from the reversetranscriptase-polymerase chain reaction (RT-PCR) prod-uct of PC-3 human prostate cancer cells (purchased fromAmerican Type Culture Collection) using the primers 3  -ACGATGAATGTGGCGATACT-5   and 5  -AGTGCAT-TCAAGGCTGTTGG-3  and a GAPDH probe (purchasedfrom American Type Culture Collection). Probes werepurified by agarose gel electrophoresis using a Qiaex gelextraction kit (Qiagen, Chatworth, CA). Each cDNA probewas radiolabeled with [  - 32 P]deoxyribonucleotidetriphosphate by the random-priming technique using theRediprime labeling system (Amersham). Subcloning of NRP-1 into pcDNA3.1 and Transfection The full-length cDNA for NRP-1 was subcloned into the Bam HI site of pcDNA3.1 (Invitrogen, San Diego, CA) aspreviously described. 27 Vectors containing NRP-1 orvector alone (pcDNA3.1) were transfected intoKM12SMLM2 cells by lipofection according to the man-ufacturer’s protocol (Hoffman-La Roche, Ltd., Basel,Switzerland). Selective medium containing 200   g/ml ofhygromycin was added 48 hours later, and viable colo-nies were selected and expanded. Cells from subconflu-ent cultures were then harvested for Western blot analy-sis and  in vivo  animal experiments as described below.  Animals and Tumor Inoculation Eight-week-old male nude mice were obtained from theNational Cancer Institute’s Animal Protection Area (Fred-erick, MD) and acclimated for 1 week while caged ingroups of five. Mice were fed a diet of animal chow andwater  ad libitum  throughout the experiment. Mice wererandomly assigned to one of three groups (10 mice pergroup); body weight at assignment was similar amongthegroups.Aftercellviabilitywasverifiedasbeing  80%by trypan blue exclusion, control and experimentalKM12SMLM2 cells (1  10 6 cells in 200   l) were injectedby means of a 30-gauge needle and 1-ml syringe sub-cutaneous in the right flank of the animals. Tumor growthwas measured every second to third day. Tumor volumewas calculated as (diameter 2   length)/2. All of the ani-mal studies were approved by the Institutional AnimalCare and Use Committee of the M.D. Anderson CancerCenter. Animals in all of the three groups were killed 18days after tumor cell inoculation because of lethargy and Neuropilin-1 in Colon Cancer 2141 AJP June 2004, Vol. 164, No. 6   the first signs of the moribund state. Tumors were har-vested and placed in either 10% formalin for paraffinfixation or OCT. Immunohistochemical Staining and Quantification of CD31 Frozen tumors from animal experiments were sectionedand processed as described above. Staining and quan-tification of CD-31 (vessels) was performed as previouslydescribed. 27 Briefly, CD-31-positive endothelial cellswere detected by localized red fluorescence with a rho-damine filter mounted on a Zeiss universal microscope(Carl Zeiss, Thornwood, NY). Tumor vessels werecounted 2 mm from the tumor edge in four distinct quad-rants at   100 magnification. Results were confirmed bytwo observers in a blinded manner. Tumor vessel areawas selectively measured in pixels 2 using NIH Image1.62 imaging software. Necrotic areas were excluded. Determination of Tumor Cell Growth To determine the effect of NRP-1 overexpression onmonolayer cell growth  in vitro , NRP-1 and pcDNA3.1empty vector transfectants were plated into 96-wellplates (Becton Dickinson, Franklin Lakes, NJ) at a densityof 2000 cells per well in 10% MEM. Cell number wasassessed by measuring the mitochondrial reduction of3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bro-mide (MTT, Sigma) to formazan. MTT was added at 24,48, and 72 hours after plating at a final concentration of2.5 mg/ml. After incubation for 90 minutes, the mediumand MTT were removed and dimethyl sulfoxide was addedfor 1 minute. Optical density was measured at 570 nm. Determination of Endothelial Cell Growth inResponse to Conditioned Media To determine the effect of conditioned media from NRP-1-transfected KM12SMLM2 colon adenocarcinoma cellson the growth of endothelial cells, NRP-1 and pcDNA3.1empty vector transfectants were grown to 80% conflu-ence in 10% MEM. The media was then changed to 1%MEM for 48 hours. This media was then collected, cen-trifuged to remove cellular debris, and given to HUVECsplated in 96-well plates  in vitro  for 24 to 48 hours. Cellnumber was then assessed by the MTT assay as de-scribed above. Determination of Tumor Cell and Endothelial Cell Migration NRP-1- and pcDNA 3.1-transfected KM12SMLM2 cellswere seeded onto hydrated 24-well migration plates(Becton Dickinson) at a density of 70,000 cells/well(membrane insert) in FBS-free MEM. MEM containing10% FBS or 10% FBS plus VEGF 165  (10 ng/ml) were thenadded to the bottom well as chemoattractants andcells were incubated for 48 hours. Nonmigrated cellswere then removed and migrated cells were fixed andstained using the Diff-Quick fixative (Dade Behring,Deerfield, IL) using the manufacturer’s instructions. Mi-grated cells were counted in five distinct areas at  100magnification.HUVECs were seeded onto hydrated 24-well migrationplates (Becton Dickinson) at a density of 40,000 cells/well(membrane insert) in FBS-free MEM. MEM containing 1%FBS or conditioned medium from NRP-1- and pcDNA 3.1(empty vector)-transfected KM12SM/LM2 cells, as de-scribed above, were then added to the bottom well aschemoattractants; cells were incubated for 6 hours. Non-migrated cells were then removed and migrated cellswere fixed and stained using the Diff-Quick fixative (DadeBehring) using the manufacturer’s instructions. Migratedcells were counted in five distinct areas at  100 magni-fication. Immunoprecipitation and Western Blot Hybridizations Human colon cancer cells were rinsed twice with ice-coldPBS and lysed with protein lysis buffer [20 mmol/L so-dium phosphate (pH 7.4), 150 mmol/L NaCl, 1% TritonX-100, 5 mmol/L ethylenediaminetetraacetic acid, 5mmol/L phenylmethyl sulfonyl fluoride, 1% aprotinin, 1  g/ml leupeptin, and 500   mol/L Na 3 VO 4 ]. In addition,resultant tumors from  in vivo  studies were homogenizedon ice and protein was extracted as described above.For Western blot hybridization, aliquots (50   g) of proteinwere subjected to electrophoresis on 8% polyacrylamidegels followed by electrotransfer to nitrocellulose mem-branes (Schleicher & Schleicher, Keene, NH). Mem-branes were blocked with 5% fat-free milk in 0.1% Tween20 in PBS. The primary antibodies used in this study were1:150 dilution of rabbit polyclonal anti-NRP-1 (SantaCruz) and 1:1000 dilutions of rabbit polyclonal anti-phos-pho-specific p44/42 MAPK (Erk 1/2) antibody, anti-phos-pho-specific P38 MAPK antibody, anti-phospho-specificAkt antibody, and anti-  -actin antibody (all from NewEngland Biolabs). The membranes were then washedand treated with the secondary antibody labeled withhorseradish peroxidase (anti-rabbit immunoglobulin fromdonkey at a 1:3000 dilution; Amersham). Protein bandswere visualized using a commercially available chemilu-minescence kit (Amersham). Before reprobing, the mem-branes were washed with stripping solution [100 mmol/L2-mercaptoethanol, 2% sodium dodecyl sulfate, and 62.5mmol/L Tris-HCl (pH 6.7)] for 30 minutes at 50°C. Forimmunoprecipitation, 500-  g aliquots of protein were in-cubated with 10   l of NRP-1 antibody (H-286, Santa CruzBiotechnology) at 4°C for 1 hour. The protein was thenprecipitated overnight at 4°C with 25   l of Protein A/Gplus Agarose (Santa Cruz Biotechnology) per the manu-facturer’s instructions. The precipitated protein was thenwashed, denatured, and used for Western blot hybridiza-tion as described above using a different NRP-1 antibody(A-12, Santa Cruz Biotechnology) at a 1:200 dilution. 2142 Parikh et al AJP June 2004, Vol. 164, No. 6   Effects of Cytokines on NRP-1 mRNA and Protein Levels in Human Colon Cancer Cell Lines To determine the effects of the cytokines EGF, IGF-1,IL-1  , and TNF-   on NRP-1 mRNA expression, humancolon cancer cells (HT29, KM12L4, and SW-480) weregrown to subconfluence in standard medium as de-scribed above, and the medium was changed to 5%FBS-containing medium overnight. Cells were then incu-bated with EGF (50 ng/ml), IGF-1 (100 ng/ml), IL-1  (10–50 ng/ml), or TNF-   (10 to 20 ng/ml) in 1% FBS-containing medium for various times ranging from 4 to 24hours. Total RNA was extracted and NRP-1 mRNA ex-pression was determined by Northern blot analysis asdescribed below. For immunoprecipitation, cells weretreated with various cytokines for 48 hours and proteinwas then harvested for NRP-1 levels as described above. Effect of Increasing Doses of EGF on NRP-1 mRNA Induction in HT29 Cells To determine the effect of increasing doses of EGF onNRP-1 mRNA expression, HT29 human adenocarcinomacells were grown to subconfluence in standard mediumas described above and the medium was changed tomedium containing 5% FBS overnight. Cells were theincubated with EGF (0, 0.01, 0.1, 1, 5, 50, and100 ng/ml)for 24 hours in medium containing 1% FBS. Total RNAwas extracted and NRP-1 mRNA expression was deter-mined by Northern blot analysis as described above. Effect of C225 of EGF-Induced NRP-1 mRNA inHT29 Cells To determine the effect of blocking the EGF receptor(EGFR) on NRP-1 expression, HT29 cells grown underconditions described above were pretreated with theEGFR monoclonal antibody C225 (10   g/ml) for 1 hour in1% FBS-containing medium, followed by the addition ofEGF (50 ng/ml) for 24 hours. Total RNA was extractedand Northern blot analysis was performed as describedabove. Determination of EGF’s Effects on Erk 1/2, Akt, and P38 Phosphorylation in HT29 Cells To determine the effect of EGF on the protein levels andphosphorylation of the signaling intermediates Erk 1/2,Akt, and P38, cells grown under the conditions describedabove were incubated with EGF (50 ng/ml) for 0, 10, 15,30, or 60 minutes in 1% FBS-containing medium, and celllysates were obtained. Phosphorylated and total proteinlevels were determined by Western blot analysis as de-scribed above. Determination of Effects of Erk1/2, Akt, and P38MAPK Inhibition on NRP-1 Induction by EGF  To determine the effects of inhibiting activation of Erk 1/2,Akt, and P38 on NRP-1 induction, HT29 cells grown un-der the conditions described above were pretreated with200 nmol/L Wortmannin, 10   mol/L U0126, or 25   mol/LSB203580 for 1 hour in 1% FBS-containing medium, fol-lowed by the addition of EGF (50 ng/ml) for 24 hours. Inpreliminary studies, doses of the signaling inhibitorsdemonstrated inhibition of the intended pathways withoutincreasing cell apoptosis. Total RNA was extracted, andNorthern blot analysis was done as described above. Statistical and Densitometric Analysis Tumor volume, tumor mass, the number of CD31 cells,and the total vessel area were compared using unpairedStudent’s  t  -tests (InStat for Macintosh; GraphPad Soft-ware, San Diego, CA).  P   0.05 was deemed significant.Densitometric analysis of autoradiographs was per-formed using NIH Image Analysis software (V1.62) fromthe National Institutes of Health (Bethesda, MD) to quan-tify the results of Northern blot analysis. Results  Immunohistochemical Staining of Colon Adenocarcinomas for NRP-1 To investigate whether NRP-1 is expressed by humancolon adenocarcinoma and/or adjacent nonmalignantmucosa, immunoperoxidase staining was performed on20 frozen human colon adenocarcinoma specimens.NRP-1 protein was expressed in all 20 adenocarcinomaspecimens studied but was not detectable in the adja-cent nonmalignant colonic mucosa in any specimen (Fig-ure 1, top). To further ascertain the srcin of NRP-1 pro-duction, immunofluorescent double staining for NRP-1and CK-22 (an epithelial cell marker), as previously de-scribed, 27 was performed on 10 colon adenocarcinomaspecimens. NRP-1 expression was localized to the ade-nocarcinoma epithelial cells in all specimens (data notshown). Detection of NRP-1 mRNA by   in Situ Hybridization To determine the expression of NRP-1 mRNA in humancolon adenocarcinoma and nonmalignant colonic mu-cosa,  in situ  hybridization was performed on 20 humancolon adenocarcinoma specimens. NRP-1 mRNA ex-pression was present in all adenocarcinoma specimensbut was not detectable in the adjacent nonmalignantcolonic mucosa (Figure 1, bottom) in agreement with theresults of immunohistochemical staining. Neuropilin-1 in Colon Cancer 2143 AJP June 2004, Vol. 164, No. 6 
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!