| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Tumor Biology |
Departments of Cancer Biology [O. S., W. L., M. F. M., J. S. W., F. F., N. R., M. K., C. D. B., L. M. E.] and Surgical Oncology [S. A. A., A. A. P., L. M. E.], The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In human colorectal cancer, Angs seem to be expressed differently in tumors and nonmalignant tissues (13) . In previous studies from our laboratory, Ang-2 was expressed ubiquitously in tumor epithelium of human colon cancer specimens, whereas expression of Ang-1 in tumor epithelium was rarely detected. This observation suggests that a net gain in Ang-2 activity over Ang-1 activity might be an initiating factor for tumor angiogenesis (14, 15, 16) .
Although several investigators have shown that loss of Ang-1 activity may augment tumor angiogenesis, others have suggested that Ang-1 is proangiogenic (17, 18, 19, 20, 21, 22, 23, 24, 25, 26) . Thus, the effects of Angs on angiogenesis and tumor growth remain controversial. In a previous study, we demonstrated that imbalances in Ang expression significantly affected angiogenesis and tumor growth of s.c. implanted colon cancer cells [HT29 (27) ]. In that study, Ang-1 overexpression inhibited angiogenesis and tumor growth of s.c. xenografts. However, the role of Ang-1 in tumor angiogenesis and growth at the site of metastatic tumor growth remains undefined. The influence of Ang-1 in the mediation of tumor vascular permeability and pericyte coverage is also poorly defined.
In this study, we hypothesized that overexpression of Ang-1 by tumor cells would impair angiogenesis and thereby inhibit tumor growth of human colon cancer cells implanted into livers of nude mice (the liver is the most common site of colon cancer metastases). We also investigated the pure angiogenic effect of a novel recombinant Ang-1 (Ang-1 TFD) in an in vivo angiogenesis assay and investigated the role of Ang-1 in vascular permeability and pericyte coverage.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Stable Transfection.
The full-length cDNA for Ang-1 was obtained from Dr. T. Gilmer (GlaxoSmithKline, Research Triangle Park, NC). The construct was subcloned into a pcDNA3.1 vector (InVitrogen, Carlsbad, CA) containing a hygromycin resistance gene. The vector containing Ang-1 or the empty pcDNA vector was transfected into HT29 cells with Lipofectin according to the manufacturer’s protocol [Boehringer Mannheim Co., Randburg, South Africa (27)
]. Cells were then grown in selective media (10% FBS-MEM containing 200 ng/ml hygromycin). Cell clones were subsequently screened by Northern blot analysis for an increase in Ang-1 mRNA expression relative to that in the pcDNA-transfected cells, as described previously (27)
. For in vivo experiments, HT29 cells that had been transfected with Ang-1 or pcDNA were harvested from subconfluent cultures by rinsing them with PBS and trypsinizing (0.25% trypsin and 0.02% EDTA) them for 3 min. Cells were washed in 10% FBS-MEM and counted. Cell viability was assessed by trypan blue exclusion, verifying that cell viability was >90% in both cell lines. Cells were then centrifuged and resuspended in HBSS for injection into mice.
Quantification of VEGF Protein in CM from Transfected Colon Cancer Cells.
VEGF protein concentrations in conditioned media from Ang-1- or pcDNA-transfected HT29 cells were determined using an ELISA kit for human VEGF (Biosource International, Camarillo, CA). Cell culture supernatants (3 ml) from cells were collected after a 48-h incubation period [in 10% FBS-MEM (FBS does not contain detectable human VEGF)]. Supernatants were subsequently collected after centrifugation for 5 min at 350 x g. In parallel, cells in culture flasks were rinsed with PBS, trypsinized as described above, resuspended in 10% FBS-MEM, and counted for each cell line. VEGF ELISA of 10-fold-diluted conditioned media (due to excessive high VEGF levels) was performed according to the manufacturer’s protocol.
Animals.
Eight-week-old male athymic nude mice or BALB/c mice (both obtained from the Animal Production Area of the National Cancer Institute and Development Center, Frederick, MD) were acclimated for 1–2 weeks while caged in groups of five. Mice were housed as described previously (29
, 30)
and fed a diet of animal chow and water ad libitum throughout the experiment. All experiments were approved by the Institutional Animal Care and Use Committee of The University of Texas M. D. Anderson Cancer Center.
Colon Cancer Liver Tumor Model.
To determine the effects of Ang-1 transfection on hepatic tumor growth of human colon cancer cells, Ang-1-transfected or pcDNA-transfected (control) cells (1 x 106 cells in 50-µl injection volume) were directly injected into the livers of athymic nude mice after they had been randomly assigned to one of the two groups (7–9 mice/group). Body weight was similar between the groups at the beginning of the experiment. Mice were observed daily, and all mice were killed (when three in any one group showed decreased mobility or discomfort) by cervical dislocation after anesthesia induction with pentobarbital (Nembutal; 50 mg/kg). Body weights were measured, livers were excised, and liver weights and tumor diameters were subsequently determined. Tumor volumes were calculated with the equation width2 x length x 0.5. Tumor tissue was then harvested and either placed in 10% formalin for paraffin embedding or snap-frozen in OCT solution (Miles Inc., Elkhart, IN) in preparation for subsequent immunohistochemical analyses.
To confirm the paracrine effect of Ang-1 on in vivo angiogenesis, Ang-1 and pcDNA cells were mixed at various ratios (100% Ang-1:0% pcDNA, 50% Ang-1:50% pcDNA, 10% Ang-1:90% pcDNA, and 0% Ang-1:100% pcDNA) and injected into the liver as described above. Mice were observed and sacrificed according to the criteria described above. Tumor-bearing livers were excised, their weights were determined, and tumor diameters were measured.
Gelfoam in Vivo Angiogenesis Assay.
Effects of Ang-1 on angiogenesis were investigated in a Gelfoam in vivo angiogenesis assay using male BALB/c mice (5 mice/group). Recombinant human Ang-1 [Ang-1 TFD (clustered Ang-1 is required for activation of Tie-2)] was the generous gift of Dr. Jocelyn Holash (Regeneron Pharmaceuticals, Tarrytown, NY). The structure and clustering of Ang-1 TFD were recently further characterized by Davis et al. (31)
. Sterile absorbable sponges (Pharmacia, Peapack, NJ) were cut into 5 x 5 x 7-mm pieces and hydrated overnight at 4°C in sterile PBS. Excess PBS was then drained by blotting onto sterile filter paper. The sponges were then soaked with 0.4% agarose (100 µl) containing either PBS (control) or Ang-1 TFD (1 µg/µl). The agarose-Gelfoam plugs were then allowed to harden for 1 h at room temperature before their s.c. implantation into BALB/c mice (5 mice/group). Mice were anesthetized with Nembutal (50 mg/kg), and the plugs were implanted s.c. via a midline incision of the abdominal skin and placed either toward the right flank (Ang-1 TFD plugs) or the left flank (PBS plugs). Fourteen days later, mice were sacrificed as described above (32)
. Gelfoam plugs were harvested, placed in OCT solution, and snap-frozen in liquid nitrogen for subsequent immunohistochemical analyses.
Immunohistochemical Analyses of Tumor Vessel Density.
Rat antimouse CD31/PECAM-1 antibody was obtained from PharMingen (San Diego, CA), and peroxidase-conjugated goat antirat IgG was obtained from Jackson ImmunoResearch Laboratories (West Grove, PA). Tumors that had been frozen in OCT were sectioned in 8-µm slices, mounted on positively charged slides, and air-dried for 30 min. Tissue sections were then fixed in cold acetone followed by 1:1 acetone/chloroform and acetone and then washed with PBS. Specimens were then incubated with 3% H2O2 in methanol for 12 min at room temperature to block endogenous peroxidase, washed three times with PBS (pH 7.5), and incubated for 20 min at room temperature in a protein-blocking solution consisting of PBS supplemented with 1% normal goat serum and 5% normal horse serum. The primary antibody directed against CD31 was diluted 1:800 in protein-blocking solution and applied to the sections, which were incubated overnight at 4°C. Sections were then rinsed in PBS and incubated for 10 min in protein-blocking solution before the addition of peroxidase-conjugated secondary antibody. The secondary antibody used for CD31 (peroxidase-conjugated goat antirat IgG) staining was diluted 1:200 in protein-blocking solution. After incubation with the secondary antibody for 1 h at room temperature, the samples were washed and incubated with stable diaminobenzidine (Research Genetics, Huntsville, AL) substrate. Staining was monitored under a bright-field microscope, and the reaction was stopped by washing with distilled water. Sections were counterstained with Gill’s No. 3 hematoxylin (Sigma Chemical Co., St. Louis, MO) and mounted with Universal Mount (Research Genetics). CD31-stained vessels were counted (at x50 magnification) at four different quadrants of each tumor (2 mm inside the tumor-normal tissue interface), and averages were calculated. For all immunohistochemical studies, the primary antibody was omitted as a negative control.
Immunohistochemical Analyses of Tumor Cell Proliferation.
Paraffin-embedded tissues were sectioned and stained for PCNA by using the mouse anti-PCNA clone PC10 DAKO A/S from DAKO (Carpinteria, CA). Paraffin-embedded tissues were sectioned in 4–6-µm slices, mounted on positively charged Superfrost slides (Fisher Scientific Co., Houston, TX), and dried overnight. Sections were deparaffinized in xylene, followed by treatment with a graded series of alcohol washes [100%, 95%, 80% ethanol/double-distilled H2O (v/v)], rehydration in PBS (pH 7.5), and microwaving for 5 min for antigen retrieval. Immunohistochemical procedures were performed as described previously (29)
. Positive reactions were visualized by incubating the slides with stable diaminobenzidine for 10–20 min. The sections were rinsed with distilled water, counterstained with Gill’s hematoxylin for 1 min, and mounted with Universal Mount (Research Genetics). Slides were also stained with H&E to study overall tissue structure. The numbers of PCNA-positive and PCNA-negative tumor cells were determined in 4 random fields/tumor (at x100 magnification), and the percentage of PCNA-positive cells was then calculated.
Immunofluorescent Analyses of Pericyte-covered Tumor Vessels and Ang-1 Expression in Hepatic Tumors.
To determine pericyte coverage of tumor vessels in pcDNA-transfected and Ang-1-transfected liver tumors, double staining for CD31 and 
-SMA (DAKO) was performed according to a modified protocol as described elsewhere (33)
. Frozen sections of hepatic tumors were stained overnight (at 4°C) for CD31/PECAM-1 (PharMingen) after acetone fixation as described above. Slides were rinsed with PBS (three times for 3 min each time) and incubated for 10 min in protein-blocking solution before the addition of Texas Red-conjugated goat antirat secondary antibody (1:200; Jackson ImmunoResearch Laboratories) and subsequent incubation for 1 h at room temperature under light protection. Antibodies were washed off with PBS (three washes for 3 min each), and slides were blocked with nonspecific goat antimouse IgG Fab fragment (Jackson ImmunoResearch Laboratories), diluted 1:10 in protein block, for 1 h at room temperature to reduce background staining for the subsequent double staining procedure. After another rinsing cycle with PBS (three times for 3 min each time), slides were incubated for 10 min in protein-blocking solution. For pericyte staining (pericytes were defined as -SMA-positive cells in direct contact with ECs), tumor sections were incubated overnight (at 4°C) with mouse anti--SMA (DAKO; 1:2000 in protein block solution). The antibody was then rinsed off with PBS, and protein-blocking solution was applied for 10 min. Alexa 488 (green; Jackson ImmunoResearch Laboratories) rabbit antimouse secondary antibody (1:200 in protein block solution) was added for 1 h at room temperature. Slides were rinsed in PBS, and nuclei were stained with Hoechst dye (1:2000) for 2 min. Slides were analyzed with an epifluorescence microscope equipped with narrow bandpass excitation filters (Chroma Technology Corp., Brattleboro, VT) to individually select for green, red, and blue fluorescence. Images were captured with a C5810 Hamamatsu camera (Hamamatsu Photonics K.K., Bridgewater, NJ) mounted on a Zeiss Axioplan microscope (Carl Zeiss Inc., Oberkochen, Germany) using Optimas image analysis software (Media Cybernetics, Silver Spring, MD). Images were further processed with Adobe Photoshop software (Adobe Systems, Mountain View, CA). Double-stained slides were analyzed at x200 magnification for the degree of pericyte/CD31 colocalization as described elsewhere (33)
. The degree of pericyte coverage was evaluated in 4 fields/tumor (2 mm inside the tumor-normal tissue interface) and rated as either absent or full coverage (defined as covering >90% of the vessel). The average percentage of covered vessels relative to uncovered vessels was then calculated for each tumor.
To confirm that Ang-1 was overexpressed in hepatic tumors, immunohistochemistry detection of Ang-1 was analyzed using goat anti-Ang-1 antibody N-18 (Santa Cruz Biotechnology, Santa Cruz, CA). Frozen tumor sections were fixed and blocked with protein block solution as described above. The primary antibody was added at 1:100 dilution and incubated overnight at 4°C. Slides were processed and analyzed as described, except that Alexa 488 antigoat antibody (1:500) was added as secondary antibody for 1 h.
Immunofluorescent Analyses of Vessel Density in Gelfoam Plugs.
Frozen sections of the agarose-Gelfoam plugs were prepared and stained for CD31/PECAM-1 (PharMingen) as described above. For immunofluorescent analysis, slides were incubated with Texas Red-conjugated goat antirat secondary antibody (1:200) as described above. Vessels were counted under an epifluorescence microscope at four different hot spots in each Gelfoam plug at x50 magnification as described in the previous paragraph.
In Vivo Miles Permeability Assay.
To investigate the effects of Ang-1 overexpression by tumor cells on vascular permeability, an intradermal Miles assay was performed. CM from Ang-1-transfected or pcDNA-transfected HT29 cells was collected after a 48-h incubation in 1% FBS-modified Eagle’s medium at 80% cell density, centrifuged for 5 min at 350 x g, and filtered through a 0.22-µm filter (Corning Inc., Corning, NY). Nude mice (n = 4) received i.v. injection with sterile 0.5% Evans blue dye (200 µl) via the tail vein. Ten min later, mice were given intradermal injections into the dorsal skin at three different sites, one for CM-Ang-1, one for CM-pcDNA, and one for VEGF (10 ng/ml; R&D Systems Inc., Minneapolis, MN). The VEGF served as positive control for increased vascular permeability. The intradermal injections (50 µl/injection) were done with a 30-gauge needle. Mice were killed 20 min after the intradermal injections by cervical dislocation after anesthesia had been induced with Nembutal. The dorsal skin of each mouse was harvested to permit visualization of intradermal dye leakage. To determine the relative degree of vascular permeability, two dimensions (a and b) of the elliptically appearing area of dye leakage were obtained at each injection site by an observer blinded to the experimental group, and the area was calculated with the formula a x b x 
.
Densitometric Quantification of Vascular Permeability.
Densitometric analysis was performed using the NIH Image Analysis software (V1.62) from the NIH (Bethesda, MD) as another means of quantifying the extent of dye leakage at the intradermal injection sites in each mouse. Digitally obtained images of the underside of the dorsal skin, including all injection sites, were converted to a gray-scale image, and dye density was analyzed at each site (the threshold was set individually for each dorsal skin flap but was constant for each mouse).
EC Coculture and Tie-2 Phosphorylation Assay.
To verify functional overexpression of Ang-1 in transfected HT29 cells, Ang-1- and pcDNA-transfected HT29 cells were cocultured with HUVECs for 48 h using transwell culture dishes (Corning Inc.). ECs were then harvested in PBS, and protein was isolated for Tie-2 immunoprecipitation as described previously (17)
. Briefly, 600 µg of protein for each experiment were used for immunoprecipitation using rabbit Tie-2 antibody (sc-324) and A/G plus agarose (both from Santa Cruz Biotechnology). Protein was then separated on a denaturating 7.5% SDS-polyacrylamide gel for Western blot analysis of Tie-2 phosphorylation. Membranes were probed with mouse anti-phosphotyrosine (Upstate Biotechnology, Lake Placid, NY), and Tie-2 phosphorylation levels were analyzed by densitometry. Membranes were additionally probed for Tie-2 to assure equal loading.
Statistical Analyses.
All statistical analyses were done using InStat Statistical Software (V2.03; GraphPad Software, San Diego, CA), with Ps of <0.05 considered to be statistically significant. Results of in vivo experiments were also tested for significant outliers using the Grubb’s test for assessing outliers.4
Tumor-associated variables were tested for statistical significance using the two-tailed Student’s t test or the Mann-Whitney U test (for nonparametric data) as specified in the figure legends. Fisher’s test was applied for comparing the incidence of hepatic tumor formation.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Ang-1 overexpression in tumors also led to a significant decrease in tumor volume (P < 0.05; Fig. 1B
). Representative photographs of excised livers are presented in Fig. 2
. To confirm that Ang-1 secretion from the experimental cell line had no direct effect on tumor cell proliferation, the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay was performed as described previously (34)
. Growth rates of both Ang-1- and pcDNA-transfected cells were similar over 24 and 48 h.
|
|
. To investigate whether Ang-1 overexpression has indirect effects on tumor cell proliferation (HT29 cells are Tie-2 negative by reverse transcription-PCR; data not shown), we evaluated the percentage of proliferating tumor cells by immunohistochemical staining for PCNA. Ang-1-overexpressing colon cancer tumors demonstrated significantly lower percentages of proliferating (PCNA-positive) tumor cells than did tumors in the pcDNA group (P < 0.01; Fig. 3B
). Representative immunohistochemical images of H&E, CD31, and PCNA analysis are demonstrated for Ang-1 and pcDNA tumors in Fig. 4
.
|
|
-SMA; green cells adjacent to ECs). Ang-1 overexpression in HT29 hepatic tumors significantly increased the degree of pericyte coverage in the Ang-1 group (P < 0.01) as compared with pcDNA tumors (Fig. 5A)
. In quantitative terms, 68% of tumor vessels in the Ang-1-transfected tumors were tightly associated with and surrounded by pericytes, but only 13% of tumor vessels in the pcDNA tumors showed the same degree of high coverage (Fig. 5B)
.
|
). Similar to our first study, cell suspensions containing only Ang-1-transfected (100%) cells formed significantly smaller tumor volumes compared with controls (100% pcDNA; P < 0.04). Liver weights in the 100% Ang-1 group were again significantly lower compared with controls (100% pcDNA; P < 0.03; data not shown).
|
).
|
).
|
2-fold increase in Tie-2 phosphorylation in HUVECs exposed to Ang-1-transfected cells, compared with Tie-2 phosphorylation levels in HUVECs cocultured with pcDNA HT29 cells (data not shown). |
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The functional complexity of Angs in the regulation of angiogenesis and in their effects on tumor growth is mirrored by conflicting reports on the in vivo effects of Ang-1 and Tie-2 activation (18, 19, 20 , 22, 23, 24 , 27 , 36) . Initially, Holash et al. (26) elegantly demonstrated the importance of coordinated induction of Angs and VEGF in tumor angiogenesis. Several studies have suggested that Ang-1 may be, in general, proangiogenic (6 , 10 , 11 , 18 , 24) . Increased neovascularization by Ang-1 was demonstrated in transgenic mouse models, where Ang-1 overexpression by keratinocytes, in combination with endogenous VEGF expression, led to increased dermal vascularization in mice, suggesting that VEGF and Ang-1 play coordinated and complementary roles. Thus, it was suggested that Ang-1 be used in combination with VEGF for promoting therapeutic angiogenesis (11 , 18) . The importance of cooperation of Ang-1 and VEGF for promotion of angiogenesis has been demonstrated in several malignant and nonmalignant models of angiogenesis (6 , 10) . However, until recently, only a few reports were available on the role of the Angs in tumor angiogenesis. In contrast to our previous report (27) on the effects of imbalances in Ang-1 and -2 expression in colon cancer cells, where Ang-1 overexpression inhibited angiogenesis and growth of xenografted tumors, Shim et al. (24) demonstrated that antisense Ang-1 mRNA expression by HeLa cervical adenocarcinoma cells inhibited angiogenesis and growth of xenografted tumors in immunodeficient mice. A different approach to Ang-1 inhibition was used by Lin et al. (19) , who demonstrated that a soluble Tie-2 receptor could decrease angiogenesis and tumor growth of murine melanoma and mammary tumors when delivered by an adenoviral vector.
Recently, several studies suggested that overexpression or administration of Ang-1 may inhibit both neoplastic and nonneoplastic angiogenesis (17 , 25 , 36, 37) . Joussen et al. (37) investigated the effects of intravitreal Ang-1 application on retinal vascularization in diabetic rats. In that study, Ang-1 decreased retinal neovascularization and normalized VEGF levels. Similar results were found when Ang-1 was delivered systemically by an adenoviral vector (37) . A blunted proangiogenic effect of VEGF by Ang-1 was also described by Visconti et al. (38) in a transgenic mouse model of cardiac-specific expression or coexpression of Ang-1, Ang-2, or VEGF. In our previous studies, we were able to demonstrate that imbalances in Ang expression may regulate growth and angiogenesis of human colon cancer. Ang-1 overexpression significantly inhibited tumor angiogenesis in that xenograft model (27) . In the present study, our results again showed that overexpression of Ang-1 significantly reduced tumor growth (79%) and neovascularization (25%), this time in a model of colorectal cancer growing at the preferred site for metastases (i.e., liver). Additionally, by using various mixtures of cell suspensions of Ang-1- and pcDNA-transfected cells, we were able to demonstrate for the first time that Ang-1 secreted by tumor cells may impact the growth of tumor cells that do not express Ang-1. In this experiment, a mixture of 10% Ang-1-transfected cells with 90% pcDNA-transfected cells was sufficient to significantly reduce tumor growth in this group compared with the 100% pcDNA group. This suggests that adding Ang-1 to the tumor microenvironment may significantly inhibit the angiogenic process, resulting in an overall inhibition of tumor growth. The antiangiogenic effect of Ang-1 observed in our study may be mediated in part by increased periendothelial support by pericytes (high pericyte coverage in Ang-1 tumors), resulting in overall vessel stabilization and thereby inhibition of initiation of tumor angiogenesis. Because HT29 cells do not express Tie-2, the observed effects of Ang-1 expression on tumor growth result from effects on ECs and periendothelial cells rather than from effects on tumor cells themselves. However, Carslon et al. (39) recently demonstrated that Ang-1 may also interact with the extracellular matrix and mediate certain effects through a Tie-2-independent mechanism that involves various integrins. It rather unlikely that this mechanism plays a significant role in our tumor model because most integrins are absent on HT29 cells. In apparent contrast to our findings are those from Stratman et al. (40) , who investigated the effects of Tie-2 inhibition in breast cancer cell lines that express various levels of Tie-2. Their results showed that transfection of a dominant-negative Tie-2 construct led to 15% growth inhibition in a Tie-2-negative cell line and 57% growth inhibition in a Tie-2 positive cell line, respectively. However, these observed effects are difficult to explain based on current data on the role of Tie-2 in mediating tumor angiogenesis (40) .
Ang-1 also inhibited nonneoplastic neovascularization, as demonstrated by our in vivo angiogenesis assay with a novel recombinant Ang-1 (Ang-1 TFD; Ref. 31
). The finding that the Gelfoam plugs were negative for -SMA expression (data not shown) suggests that Ang-1 has a direct inhibitory effect on ECs in vivo.
Our findings are supported by the results of a recent study in which Ang-1 overexpression by MCF-7 breast cancer cells resulted in stabilization of blood vessels associated with the tumor edge (25) . In that study, tumor cell proliferation decreased significantly in the presence of Ang-1 and prevented vessel dilation and dissociation of smooth muscle cells from existing vessels, which resulted in reductions in xenografted tumor growth. On the basis of results from their Matrigel in vivo assay, in which Ang-1 increased mesenchymal cell infiltration, Tian et al. (25) concluded that vascular stabilization by Ang-1 accounts for the inhibition of tumor growth. They also demonstrated that Tie-2 was expressed on smooth muscle cells in culture (25) . In a previous study, Hayes et al. (36) also demonstrated that Ang-1 overexpression in MCF-7 human breast cancer cells caused a significant retardation in tumor growth despite the high coexpression of a potent angiogenic growth factor (fibroblast growth factor-1). The same growth-inhibitory effect (70% reduction) by Ang-1 was observed by Hawighorst et al. (17) in stable transfected human squamous cell carcinoma cells. Those authors did not detect changes in vessel density but confirmed a significant increase in pericyte-covered vessels in Ang-1-transfected tumors (17) . Results from our study expand these findings, showing that Ang-1 in the tumor microenvironment may also recruit pericytes into hepatic metastasis. Sundberg et al. (41) recently described that pericytes express Ang-1 at later stages of the angiogenic process, leading to further vessel stabilization and maturation of the tumor neovascular network. Taken together, these studies suggest that continuous Tie-2 activation on ECs may lead to increased vessel stabilization, thereby making the vasculature less susceptible to proangiogenic factors such as VEGF.
Vessel stabilization by Ang-1 is associated with decreased vascular permeability. In our in vivo permeability assay, Ang-1 levels in CM from Ang-1-transfected cells abrogated the increase of plasma leakage (dye leakage) caused by tumor cell-derived growth factors. Similar results were obtained with CM from transfected KM12L4 cells [high constitutive VEGF expression (35) ], suggesting that Ang-1 is an important mediator of vascular stabilization and permeability and may override VEGF-mediated vessel leakage (42) . This phenomenon has been described by other groups who have investigated the effects of Ang-1 on vascular permeability and vessel stabilization (11 , 12) . Thurston et al. described anti-permeability properties of Ang-1 in two different studies (11 , 12) , one evaluating the effect of VEGF on plasma leakage of adult vasculature, and another with a transgenic mouse model in which both Ang-1 and VEGF were overexpressed. In the mouse study, coexpression of Ang-1 and VEGF resulted in the formation of leakage-resistant vessels (11) . The authors also showed that acute administration of Ang-1 protected adult vasculature from leakage mediated by VEGF and inflammatory cytokines (12) . The molecular mechanism of this regulatory effect was recently described by Gamble et al. (43) , who showed that administration of recombinant Ang-1 supported the localization of a cell adhesion molecule (PECAM-1) into junctions between ECs, thereby strengthening these junctions.
In conclusion, our study indicates that Ang-1 expression or administration may negatively regulate angiogenesis and decrease vascular permeability by stabilizing ECs and increasing periendothelial support. Thus, Ang-1 is an important mediator of neoplastic and nonneoplastic angiogenesis; however, its precise role in this process remains to be elucidated. Sequential expression of Ang-1, Ang-2, and VEGF has been shown to be crucial for successful angiogenesis (9 , 10) . Therefore, any interruption or disturbance in this balanced expression will probably affect the angiogenic process significantly. Such a disturbance could occur at the level of continuous Tie-2 activation (by Ang-1) or by Tie-2 interruption (soluble Tie-2, Tie-2 receptor antagonists). Results of future studies will be required to provide further insight into this complex process. Thus far, our findings suggest that Ang-1 might be useful as an antiangiogenic or anti-permeability agent in the treatment of metastatic colorectal cancer.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported in part by NIH Grant T-32 09599 (to S. A. A., J. S. W., and A. A. P.), NIH Grant U54 CA90810-01 (to L. M. E.), and a grant from The George and Barbara Bush Foundation for Innovative Cancer Research (to L. M. E.). 
2 To whom requests for reprints should be addressed, at Department of Surgical Oncology, Box 444, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030-4009. Phone: (713) 792-6926; Fax: (713) 792-4689; E-mail: lellis{at}mdanderson.org
3 The abbreviations used are: VEGF, vascular endothelial growth factor; EC, endothelial cell; Ang, angiopoietin; FBS, fetal bovine serum; HUVEC, human umbilical vein endothelial cell; TFD, tetra fibrinogen domain; OCT, optimum cutting temperature; PECAM, platelet/endothelial cell adhesion molecule; PCNA, proliferating cell nuclear antigen; -SMA, -smooth muscle actin; CM, conditioned medium.
Received 9/13/02. Accepted 4/16/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. DeWever, F. Frerart, C. Bouzin, C. Baudelet, R. Ansiaux, P. Sonveaux, B. Gallez, C. Dessy, and O. Feron Caveolin-1 Is Critical for the Maturation of Tumor Blood Vessels through the Regulation of Both Endothelial Tube Formation and Mural Cell Recruitment Am. J. Pathol., November 1, 2007; 171(5): 1619 - 1628. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Braun, H. Xu, F. Hu, P. Kocherlakota, D. Siegel, P. Chander, Z. Ungvari, A. Csiszar, M. Nedergaard, and P. Ballabh Paucity of Pericytes in Germinal Matrix Vasculature of Premature Infants J. Neurosci., October 31, 2007; 27(44): 12012 - 12024. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. S.N. Shim, I. A.W. Ho, and P. E.H. Wong Angiopoietin: A TIE(d) Balance in Tumor Angiogenesis Mol. Cancer Res., July 1, 2007; 5(7): 655 - 665. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. V. Dickson, J. B. Hamner, C. J. Streck, C. Y.C. Ng, M. B. McCarville, C. Calabrese, R. J. Gilbertson, C. F. Stewart, C. M. Wilson, M. W. Gaber, et al. Continuous Delivery of IFN-{beta} Promotes Sustained Maturation of Intratumoral Vasculature Mol. Cancer Res., June 1, 2007; 5(6): 531 - 542. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. I. Lin, J. Yu, T. Murata, and W. C. Sessa Caveolin-1-Deficient Mice Have Increased Tumor Microvascular Permeability, Angiogenesis, and Growth Cancer Res., March 15, 2007; 67(6): 2849 - 2856. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Hu, N. Ferrara, and R. B. Jaffe Paracrine VEGF/VE-Cadherin Action on Ovarian Cancer Permeability. Experimental Biology and Medicine, November 1, 2006; 231(10): 1646 - 1652. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Y. Han, C. N. Landen, J. G. Trevino, J. Halder, Y. G. Lin, A. A. Kamat, T.-J. Kim, W. M. Merritt, R. L. Coleman, D. M. Gershenson, et al. Antiangiogenic and Antitumor Effects of Src Inhibition in Ovarian Carcinoma. Cancer Res., September 1, 2006; 66(17): 8633 - 8639. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. M. Kim, K. E. Kim, G. Y. Koh, Y.-S. Ho, and K.-J. Lee Hydrogen peroxide produced by angiopoietin-1 mediates angiogenesis. Cancer Res., June 15, 2006; 66(12): 6167 - 6174. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Cascone, L. Napione, F. Maniero, G. Serini, and F. Bussolino Stable interaction between {alpha}5{beta}1 integrin and Tie2 tyrosine kinase receptor regulates endothelial cell response to Ang-1 J. Cell Biol., September 12, 2005; 170(6): 993 - 1004. [Abstract] [Full Text] [PDF]
|
|
