SEARCH

SEARCH BY CITATION

Keywords:

  • recombinant canstatin;
  • CT-26 colon carcinoma animal model;
  • human umbilical vein endothelial cell;
  • lymphatic endothelial cell;
  • angiopoietin-1

Abstract

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

We describe the effect of recombinant canstatin, the NC1 domain of the α2 chain of Type IV collagen, on suppression of angiogenesis and lymphangiogenesis both in vitro and in vivo. Recombinant canstatin produced from stably transformed Drosophila S2 cells reduced the expression of angiopoietin-1 in hypoxia mimetic agent, CoCl2-treated CT-26 cells. Recombinant canstatin inhibited proliferation, tube formation and migration of human angiopoietin-1 (rhAngpt-1)-treated human umbilical vein endothelial cells (HUVEC) and lymphatic endothelial cells (LEC). Recombinant canstatin suppressed the expression of Tie-2 and vascular endothelial growth factor-3 (VEGFR-3) transcripts in rhAngpt-1-treated HUVEC and LEC, respectively. The inhibitory effect of recombinant canstatin on tumor growth was also investigated using a heterotopic CT-26 colon carcinoma animal (BALB/c mice) model. Recombinant canstatin reduced the final volume and weight of tumors, and blood and lymphatic vessel densities of tumors, which were evaluated by CD-31 and LYVE-1 immunostaining. Immunohistochemical analysis showed that recombinant canstatin dramatically reduced the expression of angiopoietin-1 in CT-26 colon carcinoma-induced tumor, but not the expression of VEGF-C. Tie-2 and VEGFR-3 expressions were also reduced in recombinant canstatin-treated tumors. These results indicate that recombinant canstatin has anti-tumoral activities against CT-26 colon carcinoma cells. Recombinant canstatin reduces the expression of angiopoietin-1 in hypoxia-induced CT-26 cells and inhibits the angiogenic and lymphangiogenic signaling induced by angiopoietin-1. Recombinant canstatin probably inhibits angiogenesis and lymphangiogenesis via suppression of the integrin-dependent FAK signaling induced by angiopoietin-1/Tie-2 and/or VEGFR-3.

Angiogenesis, a process leading to development of new blood vessels from pre-existing vessels, plays essential roles in embryonic development, wound healing, inflammation and tumor growth.1 Angiogenesis is critical in the development and dissemination of most human tumors and is regulated by a flux of angiogenic and antiangiogenic factors within a tumor.2, 3 Micro-environmental changes caused by increased levels of angiogenic factors, such as members of the vascular endothelial growth factor (VEGF) family, the angiopoietin family and fibroblast growth factor (FGF) induce endothelial cell proliferation, migration and capillary tube formation, resulting in delivery of oxygen, nutrients and other factor to facilitate tumor growth and survival.4

Lymphangiogenesis, the generation of new lymphatic vessels from the pre-existing lymphatic system, is important in many physiological and pathological processes, including embryonic development, organ transplantation, wound healing, regeneration of tissues and organs and tumor metastasis.5 Spread of tumor cells to lymph nodes is a common occurrence in cancer and is often an early event in metastatic disease.6 Tumor lymphangiogenesis contributes to metastasis by increasing the abundance of lymphatics in and/or nearby a tumor. Tumor lymphangiogenesis is, therefore, an attractive target for therapeutics designed to restrict the metastatic spread of cancer.7 Tumor-induced lymphangiogenesis driven by tumors expressing lymphangiogenic growth factors, such as the VEGF family, FGF-2, angiopoietin-1, -2 and platelet-derived growth factors (PDGFs), is correlated with lymph node metastasis in experimental cancer models and in several types of human cancer.8, 9

Canstatin, a 24 kDa peptide derived from the C-terminal globular non-collageneous (NC1) domain of the α2 chain of Type IV collagen, is a potent inhibitor of angiogenesis with a distinct anti-tumor activity.10 Canstatin, produced from E. coli and 293 human embryonic kidney cells, dose-dependently inhibits proliferation of fetal calf serum (FCS)-stimulated human endothelial cells and induces apoptosis. Canstatin inhibits endothelial cell migration and tube formation.11 Canstatin-induced apoptosis in cultured human umbilical vein endothelial cells (HUVEC) is mediated by disruption of focal adhesion kinase (FAK)/phosphatidylinositol 3-kinase (PI3K)/Akt signaling and induction of Fas-dependent death signaling.12 In several experimental model systems, canstatin has inhibited tumor growth of prostate cancer,10 pancreatic cancer,13 breast cancer14 and lung cancer.15 Recently, we reported that recombinant canstatin produced from stably transformed Drosophila melanogaster S2 cells efficiently inhibited tumor growth in an orthotopic AT-84 oral squamous cell carcinoma animal model.16

In this work, we investigated the effect of recombinant canstatin on suppression of angiogenesis and lymphangiogenesis for in vitro experiments using CT-26 colon carcinoma cells, HUVEC and lymphatic endothelial cells (LEC). We also examined in vivo inhibitory effects of recombinant canstatin on tumor growth in a heterotopic CT-26 colon carcinoma animal model. Our findings suggest that recombinant canstatin probably inhibits angiogenesis and lymphangiogenesis via suppression of the integrin-dependent FAK signaling induced by angiopoietin-1/Tie-2 and/or VEGFR-3.

Material and Methods

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Cell lines and culture

Mouse colon carcinoma CT-26 cells were obtained from the Korean Cell Line Bank (KCLB, Korea) and maintained in RPMI-1640 (Thermo Scientific HyClone, Logan, UT) supplemented with 10% (v/v) heat-inactivated fetal bovine serum (FBS; Thermo Scientific HyClone) in a humidified incubator with 5% CO2 at 37°C. Primary HUVEC (Modern Cell and Tissue Technologies, Korea) were maintained in endothelial growth medium-2 (EGM-2; Lonza, Basel, Switzerland) containing 2% (v/v) heat-inactivated FBS in a humidified incubator of 5% CO2 at 37°C. LEC (Angiobio, Del Mar, CA) were maintained in microvascular EGM-2 (EGM-2 MV, Lonza) containing 20% (v/v) human serum (Lonza) in a humidified incubator of 5% CO2 at 37°C.

Preparation of purified recombinant canstatin

Recombinant human canstatin proteins were expressed from stably transformed Drosophila melanogaster S2 cells with a plasmid containing canstatin cDNA using the lipofectamine method.17 Recombinant canstatin from stably transformed Drosophila S2 cells was purified to homogeneity using a simple one-step Ni-NTA affinity fractionation, as described previously.17

RT-PCR analysis

CT-26 cells grown for 24 hr in 100 cm2 culture dishes (seeding density 1.0 × 106 cells/dish) were treated with different concentrations of recombinant canstatin (0, 0.5, 40 μg/ml) in the presence or absence of 100 μM CoCl2 and incubated for 24 hr. HUVEC and LEC grown for 24 hr in 100 cm2 culture dishes (seeding density 1.0 × 106 cells/dish) were treated with different concentrations of recombinant canstatin (0, 0.5, 40 μg/ml) in the presence or absence of 200 ng/ml of recombinant human angiopoietin-1 (rhAngpt-1) (R&D System Inc., Minneapolis, MN) and incubated for 24 hr. After washing with PBS, total RNAs were extracted using Trizol reagent (Invitrogen, Calsbad, CA) according to the protocol provided by the manufacturer. RNase-free DNase I (Invitrogen)-treated total RNAs were used in cDNA synthesis using an Improm-II™ Reverse Transcription System (Promega, Madison, WI). Reverse transcription was carried out according to the manufacturer's protocol in 20 μl reaction mixtures containing oligo(dT) primer. PCR products were obtained from PCR using a LA Taq polymerase kit (Takara, Japan), and 2 μl of cDNA was subjected to PCR with specific primers. PCR products were separated on 1% agarose gel and visualized under UV light after staining with ethidium bromide. PCR products were measured using TINA densitometry software Version 2.09c (Raytest, Straubenhardt, Germany).

Protein extraction and Western blot analysis

Cells were washed with PBS and lyzed with RIPA buffer (Pierce, Rockford, IL) supplemented with complete protease inhibitor cocktail tablets (Roche, Nutley, NJ). Protein extracts were collected via centrifugation at 14,000g for 20 min. Protein concentrations were determined using an RC/DC Bio-Rad assay kit (Bio-Rad, Hercules, CA) according to the manufacturer's protocol. Protein extracts were separated via 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto nitrocellulose membrane (GE Healthcare, Piscataway, NJ). Membranes were pre-incubated in blocking solution [5% (w/v) skim milk in TBS containing 0.1% Tween-20] for 1 hr, incubated with either anti-angiopoietin 1 (1:2,000 dilution in blocking solution; Santa Cruz Biotech., Santa Cruz, CA) or anti-β-actin (Santa Cruz Biotech.) overnight at 4°C, and probed with peroxidase-conjugated anti-mouse IgG (1:5,000 dilution in blocking solution; Santa Cruz Biotech.). Protein bands were detected using an enhanced chemiluminescent Western blotting detection reagent (GE Healthcare).

HUVEC and LEC proliferation assay

HUVEC [3 × 104 cells/well in endothelial basal medium (EBM-2, Lonza) containing 2% FBS] and LEC (5 × 104 cells/well in EBM-2 containing 2% FBS) were added to each well of gelatinized 24-well plates. After adding rhAngpt-1 (0, 50, 100, 200 ng/ml) and/or recombinant canstatin, cells were incubated for 48 hr. The cells were trypsinized and counted using a hemocytometer. To ensure that the observed inhibition was not due to detachment of cells from the plates, all wells were examined under an inverted microscope for evidence of cell detachment prior to cell counting.

HUVEC and LEC tube formation assay

The HUVEC and LEC tube formation assay procedure was as follows: 150 μl of a 1:1 (v/v) mixture of EBM-2 and Matrigel (BD Biosciences) was added to each well of a 48-well plate and allowed to polymerize at 37°C. HUVEC (4 × 104 cells/well) and LEC (6 × 104 cells/well) in 0.5 ml of EBM-2 containing 2% (v/v) FBS, rhAngpt-1 (200 ng/ml) and different concentrations (0, 0.5, 40 μg/ml) of recombinant canstatin were added to each well. After 8 hr, cells were imaged under a phase contrast inverted microscope using a digital camera and total tube lengths were quantified using the Image J program (NIH, MD).

HUVEC and LEC migration assay

A migration assay was performed using transwell 24-well plates and inserts with 8.0 μm pore-sized polycarbonate membranes (Corning, Corning, NY) coated with 0.1% gelatin in PBS for 1 hr at 37°C. HUVEC (3 × 104 cells/well) and LEC (5 × 104 cells/well) in EBM-2 with different concentrations (0, 0.5, 40 μg/ml) of recombinant canstatin were added to the upper chamber of the transwell insert. EBM-2 containing rhAngpt-1 (200 ng/ml) was added to the lower chamber to induce cell migration. After 24 hr at 37°C, cells on the top surface of the membranes were wiped off with cotton balls, and the cells that migrated on the underside of inserts were fixed with methanol and stained with Harris hematoxylin solution (Sigma). Five different digital images were taken per well, and the number of migrated cells was counted. Each sample was assayed in duplicate, and the experiment was repeated twice.

In vivo tumor studies

Five-week old male BALB/c mice were purchased from Orient Bio Inc. (Seongnam, Korea). Mice were provided with water and food ad libitum, and quarantined in a specific pathogen free environment with a 12 hr light and 12 hr dark photoperiod in an animal care facility accredited by the Kyung Hee University Institutional Animal Care and Use Committee. Animal care and experimental procedures followed the Kyung Hee University guidelines for the care and use of laboratory animals.

To establish a heterotopic colon carcinoma animal model, 5 × 105 CT-26 cells in 200 μl were injected into the right flank of BALB/c mice. Tumors grew for 6 days to form visible masses, after which animals were divided into groups of six mice each. Each group was treated daily with a peritumor injection of either recombinant canstatin (5, 10 mg/kg/day in PBS) or PBS (control) for 12 days. All mice were sacrificed 27 days after tumor inoculation and the tumors were excised and weighted. Tumor length and width were measured using a caliper, and the tumor volume was calculated using the standard formula [length × width squared × 0.5].18

Immunohistochemistry

Tumor specimens were immediately removed from sacrificed mice and prepared for immunohistological examination. Tumors were fixed in 10% (v/v) neutral buffered formalin overnight, embedded in paraffin and sectioned to a 5 μm thickness. Tumor sections were deparaffinized via immersion in xylene, dehydrated in a graded series of ethanol, and washed with distilled water. Thereafter, tumor sections were boiled in 10 mM sodium citrate buffer (pH 6.0) for 10 min and cooled at room temperature. To inhibit endogenous peroxidase activity, tumor sections were incubated with methanol containing 1% (v/v) hydrogen peroxide for 10 min. Tumor sections were then blocked with blocking solution [10% (v/v) normal goat serum, Dako, Glostrup, Denmark] for 1 h and incubated overnight in primary antibodies [anti-CD31, anti-LYVE-1 (Abcam, Cambridge, UK), anti-angiopoietin-1, anti-VEGF-C, anti-Tie-2, anti-VEGFR-3 (Santa Cruz Biotech)] diluted with the blocking solution. Tumor sections were probed with peroxidase-conjugated secondary antibodies and incubated with a peroxidase substrate solution (Vector Laboratory, Burlingame, CA) until the desired stain intensity developed. After counterstaining with Harris hematoxylin, tumor sections were examined under an inverted microscope (BX21; Olympus, Japan). To analyze immunohistochemical signals within the specimens, all tumor sections were digitized under a ×400 objective magnification and images were captured. Captured images were analyzed with the Image J program. The numbers of blood or lymphatic vessels per mm2 were recorded as the blood or lymphatic vessel density.

Statistical analysis

All data are presented as mean ± S.D. or S.E. Student's t test was used to compare different data groups (*p < 0.05, **p < 0.01, ***p < 0.001).

Results

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Recombinant canstatin reduced the expression of angiopoietin-1 in CoCl2-treated CT-26 cells

To analyze the effect of recombinant canstatin on the expression of the angiopoietin family genes, total RNAs were prepared from CoCl2-treated CT-26 cells in the presence or absence of recombinant canstatin. RT-PCR measurements using β-actin as an internal control showed that the level of the angiopoietin-1 transcript was increased 4.4 times in CoCl2-treated CT-26 cells (Figs. 1a and 1b). Increased expression of the angiopoietin-1 transcript due to CoCl2 was reduced by 29.3 and 53.1% in 0.5 and 40 μg/ml recombinant canstatin-treated cells, respectively. Expressions of angiopoietin-3 and -4 were also slightly increased due to CoCl2, and also decreased by the presence of recombinant canstatin. Expression of the angiopoietin-2 transcript was decreased by 17.5% in CoCl2-treated cells. The presence of recombinant canstatin further reduced the expression of the angiopoietin-2 transcript.

thumbnail image

Figure 1. Recombinant canstatin inhibited the expression of angiopoietin-1 in CoCl2-treated CT-26 cells. (a) CT-26 cells were treated with different concentrations of recombinant canstatin (0, 0.5, 40 μg/ml) in the presence or absence of 100 μM CoCl2, and incubated for 24 hr. cDNAs were generated from DNase I-treated total RNAs, and PCR was performed with specific primers for angiopoietins and β-actin. (b) The PCR products from three independent experiments in (a) were quantified and are represented as a bar diagram. The level of the angiopoietin transcripts in the control (recombinant canstatin- and CoCl2-nontreated cells) was estimated as 100%. (c, d) Angiopoietin-1 present in the intracellular (c) and medium fractions (d) was determined using Western blot analysis with anti-angiopoietin-1. (e, f) The amounts of angiopoietin-1 obtained in three independent experiments of (c) and (d) were quantified and are represented as bar diagrams. The level of angiopoietin in the control (recombinant canstatin- and CoCl2-nontreated cells) was estimated as 100%. Data are presented as mean ± S.D. of three independent experiments (*p < 0.05).

Download figure to PowerPoint

Expression of angiopoietin-1 in CoCl2-treated CT-26 cells was further confirmed by Western blot analysis. The levels of angiopoietin-1 in the intracellular and medium fractions of CoCl2-treated CT-26 cells were 1.4 and 4 times higher, respectively, than levels in CoCl2-nontreated cells. Recombinant canstatin reduced the levels of angiopoietin-1 in the intracellular and medium fractions of CoCl2-treated cells. The levels of angiopoietin-1 in the medium fractions of CoCl2-treated cells were reduced by 35.5 and 51.3%, respectively, in the presence of 0.5 and 40 μg/ml recombinant canstatin (Figs. 1d and 1f). To confirm that the protein levels in each medium fraction were the same, protein samples on SDS-polyacrylamide gel were evaluated using silver staining (data not shown). Our results indicate that expression of angiopoietin-1 in CT-26 cells is increased in response to hypoxia, and recombinant canstatin decreases the expression of angiopoietin-1.

Recombinant canstatin inhibits the proliferation of rhAngpt-1-treated HUVEC and LEC

To determine the effect of angiopoietin-1 on proliferation of HUVEC and LEC, cells were incubated for 48 hr in EBM-2 containing different concentrations of rhAngpt-1 and cell densities were measured after trypsinization (Figs. 2a and 2c). The proliferations of HUVEC and LEC were dose-dependently enhanced by rhAngpt-1. Cell densities of HUVEC and LEC cultured for 48 hr in the presence of 50–200 ng/ml rhAngpt-1 were increased by 28.7–72.2 and 80–146%, respectively, compared to angiopoietin-1-nontreated cells.

thumbnail image

Figure 2. Recombinant canstatin inhibited the proliferation of rhAngpt-1-treated HUVEC and LEC. (a, c) HUVEC (a) and LEC (c) in EBM-2 containing 2% FBS were added to gelatinized 24-well plates and treated with different concentrations (0, 50, 100, 200 ng/ml) of rhAngpt-1. After 48 hr of incubation, the cells were trypsinized and counted using a hemocytometer. (b, d) HUVEC (b) and LEC (d) were treated with different concentrations (0, 0.5, 40 μg/ml) of recombinant canstatin in the presence or absence of 200 ng/ml of rhAngpt-1 and incubated for 48 hr. The cells were trypsinized and counted using a hemocytometer. The cell densities obtained from three independent experiments are represented as bar diagrams. Data are presented as mean ± S.D. of three independent experiments (*p < 0.05, **p < 0.01, ***p < 0.001).

Download figure to PowerPoint

The proliferation of HUVEC and LEC stimulated by rhAngpt-1 was further examined in the presence and absence of recombinant canstatin (Figs. 2b and 2d). Whereas cell density for 200 ng/ml rhAngpt-1-treated HUVEC in the absence of recombinant canstatin was increased by 49.4%, cell densities for 200 ng/ml rhAngpt-1-treated HUVEC in the presence of 0.5 or 40 μg/ml of recombinant canstatin were increased by 29.7 or 5.6%, compared to rhAngpt-1-nontreated HUVEC. This result indicates that in rhAngpt-1-treated HUVEC, the increase of proliferation induced by rhAngpt-1 was diminished to 39.9 and 88.7%, respectively, by the presence of 0.5 and 40 μg/ml recombinant canstatin.

The cell density of LEC treated with rhAngpt-1 alone was increased by 78.6%. However, cell densities of LEC treated with 200 ng/ml rhAngpt-1 and 0.5 or 40 μg/ml recombinant canstatin were increased by 45.9 and 39.1%, respectively, which correspond to, respectively, 41.6 and 50.3% decreases, compared to the control (rhAngpt-1-treated and canstatin-nontreated LEC). These results indicate that rhAngpt-1 enhances proliferation of HUVEC and LEC, and recombinant canstatin reduces the proliferation of HUVEC and LEC that is stimulated by rhAngpt-1.

Recombinant canstatin inhibits tube formation and migration of rhAngpt-1-treated HUVEC and LEC

The effect of recombinant canstatin on tube formation in rhAngpt-1-treated HUVEC and LEC was determined using Matrigel-precoated 48-well plates. After 8 hr of incubation in a culture medium containing 200 ng/ml rhAngpt-1, tube formation in HUVEC and LEC was increased by 36 and 55.8%, respectively, compared to rhAngpt-1-nontreated HUVEC and LEC. In the presence of 0.5 and 40 μg/ml recombinant canstatin, tube formation in rhAngpt-1-treated HUVEC was decreased by 23.1 and 47.8% (Figs. 3a and 3b). Tube formation in rhAngpt-1-treated LEC was also decreased by 37.9 and 67.9% in the presence of 0.5 and 40 μg/ml recombinant canstatin (Figs. 3c and 3d).

thumbnail image

Figure 3. Recombinant canstatin inhibited the tube formation and migration in rhAngpt-1-treated HUVEC and LEC. (a, c) HUVEC (a) and LEC (c) in EBM-2 containing 2% FBS were added to Matrigel-precoated 48-well plates and treated with different concentrations (0, 0.5, 40 μg/ml) of recombinant canstatin in the presence or absence of 200 ng/ml of rhAngpt-1. After 8 hr, cells were imaged under a phase contrast inverted microscope using a digital camera (scale bar = 200 μm). (b, d) Total tube lengths of HUVEC (a) and LEC (b) were quantified using the Image J program. (e, g) HUVEC (e) and LEC (g) in EBM-2 containing different concentrations (0, 0.5, 40 μg/ml) of recombinant canstatin were added to the upper chamber of a transwell insert with 8.0-μm pore-sized polycarbonate membranes that were coated with 0.1% gelatin. EBM-2 containing rhAngpt-1 (200 ng/ml) was added to the lower chamber to induce cell migration. After 24 hr of incubation, cells that migrated on the underside of inserts were fixed with methanol, stained with Harris hematoxylin solution, and imaged under a phase contrast inverted microscope using a digital camera (scale bar = 200 μm). (f, h) Five digital images per well for (e) and (g) were obtained, and the numbers of migrated HUVEC and LEC were counted. Each sample was assayed in duplicate, and the experiment was repeated twice. The numbers of migrated HUVEC (f) and LEC (h) present in 320 mm2 are presented as bar diagrams. Data are presented as mean ± S.D. of three independent experiments (*p < 0.05, **p < 0.01, ***p < 0.001).

Download figure to PowerPoint

The effect of recombinant canstatin on migration of rhAngpt-1-treated HUVEC and LEC was also determined using transwell 24-well plates and inserts with 8.0 μm pore-sized, gelatin-coated polycarbonate membranes (Figs. 3e3h). After 24 hr of incubation in the presence of 200 ng/ml rhAngpt-1, the numbers of HUVEC and LEC that migrated through the polycarbonate membranes were increased by 273 and 80.9%, respectively. The migration of HUVEC and LEC that was stimulated by rhAngpt-1 was dramatically reduced by the presence of recombinant canstatin. Levels of 0.5 and 40 μg/ml of recombinant canstatin reduced the migration of HUVEC by 65.5 and 70.3%, respectively (Fig. 3f). Also, 0.5 and 40 μg/ml of recombinant canstatin reduced the migration of LEC by 47.9 and 70.1%, respectively (Fig. 3h). In the absence of rhAngpt-1, the basal tube formation and migration activities of HUVEC and LEC were reduced by the presence of recombinant canstatin. However, the proliferation inhibition of HUVEC and LEC due to recombinant canstatin was not observed under tube formation and migration assay conditions (Supporting Information Figs. 1b and 2b). These results indicate that recombinant canstatin reduces tube formation and migration in rhAngpt-1-stimulated HUVEC and LEC.

Recombinant canstatin reduces the expressions of Tie-2 and VEGFR-3, and the phosphorylation of FAK, in rhAngpt-1-treated HUVEC and LEC

Expression of the VEGF family protein receptors VEGFR-1, -2, -3 and the angiopoietin family protein receptors Tie-1, -2 in rhAngpt-1-treated HUVEC and LEC was determined using RT-PCR (Figs. 4a and 4b). Total RNAs were prepared from HUVEC and LEC treated with 200 ng/ml rhAngpt-1 in the presence and absence of recombinant canstatin. RT-PCR measurements using β-actin as an internal control showed that levels of VEGFR-1, -2,- 3, and Tie-2 transcripts were increased in rhAngpt-1-treated HUVEC. The most dramatic increases were observed for the VEGFR-3 and Tie-2 transcripts. Recombinant canstatin reduced the expression of the Tie-2 transcript, but not the VEGFR-1, -2 and -3 transcripts (Fig. 4a). The VEGFR-3 transcript was only increased in rhAngpt-1-treated LEC. Recombinant canstatin reduced the expression of VEGFR-3 in rhAngpt-1-treated LEC (Fig. 4b). Changes in the VEGFR-1, -2, Tie-1 and -2 transcript levels were not observed in rhAngpt-1- and/or recombinant canstatin-treated LEC. We also analyzed expression levels of Tie-2 and VEGFR-3 in HUVEC and LEC using Western blotting (Supporting Information Figs. 3a and 3b). Recombinant canstatin inhibited the expressions of Tie-2 and VEGFR-3 at the protein level in rhAngpt-1-stimulated HUVEC and LEC, respectively. These results hint that recombinant canstatin inhibits the expressions of Tie-2 and VEGFR-3 that are stimulated by rhAngpt-1 in HUVEC and LEC, respectively.

thumbnail image

Figure 4. Recombinant canstatin reduced the expression of the Tie-2 and/or the VEGFR-3 transcripts in rhAngpt-1-treated HUVEC and LEC. (a,b) HUVEC (a) and LEC (b) were treated with different concentrations of recombinant canstatin (0, 0.5, 40 μg/ml) in the presence or absence of 200 ng/ml of rhAngpt-1. DNase I-treated total RNAs were prepared 24 hr after incubation and RT-PCR was performed. (c, d) The amount of PCR products from three independent experiments in (a) and (b) were quantified and are represented as bar diagrams. The levels of the VEGFR-3 and Tie-2 transcripts of recombinant canstatin- and rhAngpt-1-nontreated cells were estimated as 100%. Data are presented as mean ± S.D. of three independent experiments (*p < 0.05, **p < 0.01, ***p < 0.001).

Download figure to PowerPoint

Panka and Meier12 have shown that canstatin inhibits the phosphorylation of FAK in HUVEC when exposed to endothelial cell growth factor (ECGF). To determine the inhibitory effect of canstatin on rhAngpt-1-induced FAK activation in HUVEC and LEC, cell lyzates were analyzed using Western blotting for phosphorylation of FAK. Addition of rhAngpt-1 resulted in an increase in FAK phosphorylation in HUVEC and LEC. This effect was slightly inhibited by canstatin at a level of 0.5 μg/ml. However, the effect was markedly blocked by canstatin at 40 μg/ml (Supporting Information Figs. 3a and 3b). In an additional experiment using antibodies blocking αvβ3 integrin, phosphorylation of FAK was substantially reduced by the antibody against integrin αvβ3 (anti-αvβ3) in rhAngpt-1-stimulated HUVEC and LEC (Supporting Information Figs. 3c and 3d). It is worthwhile to note that canstatin suppresses the integrin-dependent FAK signaling that is induced by rhAngpt-1in HUVEC and LEC. Taken together, our results suggest that recombinant canstatin reduces expressions of Tie-2 and VEGFR-3, and the phosphorylation of FAK in rhAngpt-1-treated HUVEC and LEC.

Recombinant canstatin inhibits tumor growth in a CT-26 colon carcinoma animal model

The anti-tumor activity of recombinant canstatin was investigated in a CT-26 colon carcinoma animal model using BALB/c mice. In a control group (with PBS), tumors grew rapidly and reached an average volume of 3,338.7 ± 322.6 mm3 (mean ± S.E.) by day 27 after inoculation with CT-26 cells. The size of the primary tumor (1,620.4 ± 343.6 mm3, 1,415.6 ± 209.0 mm3) in 5 and 10 mg/kg/day recombinant canstatin-treated animals was reduced to 48.5% and 42.4%, respectively, of the control group tumor size at 27 days (Fig. 5a). Similarly, the tumor weight for the 5 and 10 mg/kg/day recombinant canstatin-treated group was reduced to 65.9% and 48.9% of the control group weight, respectively (Fig. 5b).

thumbnail image

Figure 5. Recombinant canstatin reduced the final volume and weight of tumors, and the blood and lymphatic vessel densities of tumors. (a, b) CT-26 cells (5 × 105 cells in 200 μl PBS) were injected into the right flank of BALB/c mice. Six days after CT-26 inoculation, mice were treated daily with a peritumor injection of either recombinant canstatin (5 or 10 mg/kg/day in PBS) or PBS for 12 days. All mice were sacrificed 27 days after CT-26 inoculation, and the tumor volume (a) and tumor weight (b) were measured. (c, d) The blood and lymphatic vessels in tumor sections for each group were determined using immunohistochemical analysis with anti-CD31 and anti-LYVE-1, respectively. The number of blood vessels per mm2 is expressed as the blood vessel density (c). The number of lymphatic vessels per mm2 is expressed as the lymphatic vessel density (d). The blood and lymphatic vessel densities of the control group were estimated as 100%. Data are presented as mean ± S.E. (*p < 0.05, **p < 0.01, ***p < 0.001).

Download figure to PowerPoint

Capillaries of heterotopic tumors were identified using immunohistochemical CD31 staining for blood vessels. Tumor specimens of the control group showed an average blood vessel density of 4.41 ± 0.2 per mm2 (mean ± S.E.). In contrast, the blood vessel densities of the 5 and 10 mg/kg/day recombinant canstatin-treated groups were 1.75 ± 0.82 and 1.38 ± 0.42 per mm2, respectively, which corresponded to 39.6 and 31.5% of the control group density (Fig. 5c). Tumors were also analyzed for a marker of lymphangiogenesis using immunohistochemical staining for LYVE-1. Tumor specimens of the control group showed an average lymphatic vessel density of 3.47 ± 0.77 per mm2 (mean ± S.E.). In contrast, the lymphatic vessel densities of the 5 and 10 mg/kg/day recombinant canstatin-treated groups were 1.47 ± 0.39 and 1.35 ± 0.36 per mm2, respectively, which corresponded to 42.5 and 39% of the control group density (Fig. 5d). These results indicate that recombinant canstatin efficiently inhibits tumor growth in the CT-26 colon carcinoma animal model. Suppression of tumor growth due to recombinant canstatin could be caused by inhibition of angiogenesis and lymphangiogenesis.

Recombinant canstatin reduced the expressions of angiopoietin-1, Tie-2 and VEGFR-3 in CT-26 colon carcinoma-induced tumors

Immunohistochemical analysis was performed to evaluate the effect of recombinant canstatin on expressions of angiopoietin-1 and VEGF-C in CT-26 colon carcinoma-induced tumors. Expression of angiopoietin-1 was decreased by 54.7% in recombinant canstatin-treated tumor sections (Figs. 6a and 6b). However, expression of VEGF-C in recombinant canstatin-treated tumor sections was similar to that in PBS-treated tumor sections. Tie-2 and VEGFR-3 expressions were also determined using immunohistochemical analysis (Figs. 6c and 6d). They were decreased by 90.4 and 58.6%, respectively, in recombinant canstatin-treated tumor sections, compared to PBS-treated tumor sections. These results indicate that recombinant canstatin remarkably reduces the expressions of angiopoietin-1, Tie-2 and VEGFR-3 in CT-26 colon carcinoma-induced tumors.

thumbnail image

Figure 6. Recombinant canstatin reduced the expressions of angiopoietin-1, Tie-2 and the VEGFR-3 in CT-26 colon carcinoma-induced tumors. (a, c) The presence of angiopoietin-1, VEGF-C (a), Tie-2, and VEGFR-3 (c) in PBS-treated or 10 mg/ml/kg/day recombinant canstatin-treated CT-26 colon carcinoma tumor sections was determined using immunohistochemical analysis. All tumor sections were digitized and images were captured under a 400× objective magnification (scale bar = 100 μm). (b, d) Immunohistochemical intensities of angiopoietin-1, VEGF-C (b), Tie-2 and VEGFR-3 (d) from captured images were analyzed using the Image J program and are represented as bar diagrams. Data are presented as mean ± S.E. (*p < 0.05).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Tumor-induced angiogenesis and lymphangiogenesis are facilitated by angiogenic and lymphangiogenic growth factors, such as the VEGF family, the angiopoietin family, FGF-2, PDGFs and other factors.2, 8 Firstly, we evaluated expression of the angiopoietin family genes in CoCl2-treated CT-26 cells. The hypoxia mimetic agent CoCl2 allows preferential growth of hypoxic carcinoma cells over control cells and results in higher cell numbers.19 Sensitivity to CoCl2 is cell type specific and mild hypoxia and non-toxic concentrations of CoCl2 induce differentiation of human cancer cells with an increase in hypoxia-inducible factor-1α (HIF-1α) protein stabilizing p53.20, 21 The viability and the migration of CT-26 cells were increased by 100 μM CoCl2 and decreased by the presence of recombinant canstatin (Supporting Information Fig. 4). These results indicate that recombinant canstatin modulates the viability and migration of hypoxic CT-26 cells.

Levels of the angiopoietin-1 transcript and the angiopoietin-1 protein present in intracellular and medium fractions were increased in 100 μM CoCl2-treated CT-26 cells, but this increase was reduced by the presence of recombinant canstatin (Fig. 1). However, other changes in angiopoietin-2, -3, -4 (Fig. 1) and VEGF family transcript levels were not observed (data not shown). This strongly indicates that angiopoietin-1 is a main angiogenic and lymphangiogenic factor in CT-26 cells in response to hypoxia.

Angiopoietin-1, a ligand for receptor tyrosine kinase Tie-2, plays an essential role in remodeling and maturation of blood and lymphatic vessels.22–24 Angiopoietin-1/Tie-2 signaling induces autophosphorylation of Tie-2 and promotes vascular stability and integrity. Recombinant angiopoietin-1 (rhAngpt-1) dose-dependently enhanced proliferation of HUVEC and LEC, but recombinant canstatin reduced the proliferation of HUVEC and LEC that was stimulated by rhAngpt-1 (Figs. 2b and 2d). Canstatin induces the apoptosis in HUVEC, which is mediated by a disruption of FAK/PI3K/Akt signaling and an induction of Fas-dependent death signaling.12 Recombinant canstatin alone inhibited proliferation of HUVEC and LEC in dose-dependent manner (Supporting Information Figs. 1c and 2c). We used concentrations of 0.5 and 40 μg/ml of recombinant canstatin to show mild and moderate cell proliferation inhibition. The rhAngpt-1-stimulated proliferation was decreased by 39.9 and 41.6%, respectively, in 0.5 μg/ml recombinant canstatin-treated HUVEC and LEC. Also, the fact that the decrease in proliferation in HUVEC and LEC treated with both rhAngpt-1 and recombinant canstatin was more than the decrease in HUVEC and LEC treated with only recombinant canstatin indicates that recombinant canstatin inhibits the signaling by angiopoietin-1 in HUVEC and LEC that stimulates proliferation. Furthermore, both the tube formation and migration stimulated by rhAngpt-1 in HUVEC and LEC were remarkably inhibited by the presence of recombinant canstatin, even under conditions that did not induce HUVEC and LEC cell death (Fig. 3). The transcript levels of VEGF and the angiopoietin family protein receptors VEGFR-1, -2, -3 and Tie-2 were increased in rhAngpt-1-treated HUVEC. Although recombinant canstatin reduced the expression of VEGFR-3, the most distinct reduction due to recombinant canstatin was observed for the Tie-2 transcript (Figs. 4a and 4c). However, in rhAngpt-1-treated LEC, a significant increase due to rhAngpt-1 was only observed for the VEGFR-3 transcript, which was decreased due to recombinant canstatin (Figs. 4b and 4d). Also, recombinant canstatin inhibited the expressions of Tie-2 and VEGFR-3 at the protein level in rhAngpt-1-stimulated HUVEC and LEC, respectively. These results strongly indicate that recombinant canstatin inhibits the HUVEC and LEC proliferation, tube formation and migration that are stimulated by angiopoietin-1. Furthermore, these results suggest that recombinant canstatin inhibits angiogenesis and lymphangiogenesis that are stimulated by angiopoietin-1, which are mediated by suppression of Tie-2 and VEGFR-3 at the mRNA and protein levels in HUVEC and LEC, respectively.

We investigated the anti-tumor activity of recombinant canstatin in a heterotopic CT-26 colon carcinoma animal model using BALB/c mice. Recombinant canstatin reduced the tumor size and weight, as observed in our AT-84 oral squamous carcinoma animal.16 This finding is, to the best of our knowledge, the first report that recombinant canstatin inhibits tumor growth in a CT-26 colon carcinoma model. Recombinant canstatin reduced the blood and lymphatic vessel densities, indicating that recombinant canstatin has anti-angiogenic and anti-lymphangiogenic activities in a CT-26 colon carcinoma model. Expression of angiopoietin-1 in CT-26 colon carcinoma-induced tumors was reduced by treatment with recombinant canstatin (Figs. 6a and 6b), which also reduced expressions of Tie-2 and VEGFR-3 in tumor sections (Figs. 6c and 6d). These results indicate that recombinant canstatin reduces the tumor growth induced by CT-26 colon carcinoma, which was mediated by a decrease in angiopoietin-1, Tie-2 and VEGFR-3 levels.

Angiopoietin-1/Tie-2 signaling regulates both the maintenance of vascular quiescence and the promotion of angiogenesis that are regulated by trans-associated Tie-2 and extracellular matrix (ECM)-anchored Tie-2, respectively.23 Angiopoietin-1/ECM-anchored Tie-2 signaling promotion of angiogenesis induces the integrin-dependent focal complex formation that leads to activation of FAK, which is involved in integrin-mediated ERK1/2 activation. Integrin, a heterodimeric transmembrane glycoprotein consisting of α and β subunits that mediate cell–cell and cell-ECM connections, participates in blood and lymphatic vessel growth by promoting endothelial cell migration and survival.25, 26 Integrins αvβ3 and αvβ5 are highly linked with blood vessel development. Integrin αvβ3, expressed in proliferating endothelial cells, is a key mediator in capillary formation.27 Canstatin has been reported to trigger a crucial mitochondrial apoptotic mechanism in endothelial and tumor cells that is mediated through interaction with integrins αvβ3 and αvβ5.28 Canstatin also inhibits phosphorylation of FAK in HUVEC exposed to ECGF.12 In our experiments, recombinant canstatin reduced phosphorylation of FAK in rhAngpt-1-stimulated HUVEC and LEC. Also, phosphorylation of FAK was substantially blocked by the antibody against integrin αvβ3 (anti-αvβ3) in both rhAngpt-1-stimulated HUVEC and LEC. These results indicate that the inhibition of angiopoietin-1/Tie-2 signaling (proliferation, tube formation and migration) due to recombinant canstatin, shown here, may be caused by suppression of integrin-dependent FAK signaling. In contrast to the role of the integrins in angiogenesis, less is known about the role of integrins in LEC and their function in lymphangiogenesis. Although integrins α1β1 and α2β1 are involved in lymphangiogenesis in response to VEGF-A and integrin α4β1 is expressed on tumor and growth factor-induced lymphatic endothelium,29, 30 integrins involved in lymphangiogenesis by angiopoietin-1 have not determined yet. Angiopoietin-1 has been reported to induce lymphatic vessel enlargement, sprouting and proliferation in a VEGFR-3-dependent manner.24, 31 We have shown that recombinant canstatin reduces the level of the VEGFR-3 transcript that is increased by angiopoietin-1. These results indicate that canstatin inhibits the lymphangiogenesis promoted by angiopoietin-1 through reduction of the VEGFR-3 transcript level, but not the Tie-2 level.

In summary, expression of angiopoietin-1 was increased in hypoxia-induced CT-26 cells and recombinant canstatin decreased the expression of angiopoietin-1. Recombinant canstatin inhibited the HUVEC and LEC proliferation, tube formation and migration that were stimulated by angiopoietin-1. Recombinant canstatin reduced expressions of the Tie-2 and VEGFR-3 at both the mRNA and protein levels in angiopoietin-1-treated HUVEC and LEC. Recombinant canstatin inhibited growth of CT-26 colon carcinoma in vivo and reduced the blood and lymphatic vessel densities of tumors. Expressions of angiopoietin-1, Tie-2 and VEGFR-3 were reduced in recombinant canstatin-treated tumors. These results indicate that recombinant canstatin has anti-tumoral activities against CT-26 colon carcinoma cells and inhibits the angiogenic and lymphangiogenic signaling that is stimulated by angiopoietin-1. Recombinant canstatin probably inhibits angiogenesis and lymphangiogenesis via suppression of the integrin-dependent FAK signaling that is induced by angiopoietin-1/Tie-2 and VEGFR-3 in HUVEC and LEC.

References

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IJC_26353_sm_SuppFigs.pdf473KSupporting Figures.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.