| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Department of Pathology and Molecular Medicine, Centre for Gene Therapeutics, McMaster University, Hamilton, Ontario, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(IFN-).3,4
More recently, two powerful inhibitors of
angiogenesis, angiostatin and endostatin have been
reported.5,6,7
Angiostatin, a fragment of the plasma
protein plasminogen was initially isolated from urine of mice with
concomitant Lewis lung carcinoma.5 Previous studies have used repeated subcutaneous injections of angiostatin protein to demonstrate function,5 however the specific transient targeting of the angiostatin gene to a tumor site may be an attractive alternative using recombinant adenoviral vectors. This would allow for short term, concentrated expression of angiostatin in the tumor environment. This study demonstrates the construction and characterization of a recombinant, replication deficient type-5 adenovirus vector expressing the cDNA-encoding amino acids 98–458 of murine plasminogen. The biological activity is shown both in vitro and in vivo using human umbilical cord vein endothelial cells (HUVEC) and an in vivo Matrigel bioassay respectively. We also show that sustained expression of murine angiostatin by adenovirus delivered directly into the lung results in expression of angiostatin for up to seven days within the lung and significantly delayed the growth of metastatic breast tumor cells that had been embolized to the lung.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Six- to eight-week-old C57BL/6 and FVB/n female mice were obtained from Charles River (Troy, NY) and housed in a specific pathogen-free environment until use. All animal work was approved by the McMaster University animal ethics research board. Transgenic FVB mice expressing the polyomavirus (PyV) Middle T antigen under the transcriptional control of the mammary tumor virus (MMTV) long terminal repeat, spontaneously develop adenocarcinomas of the mammary epithelium in 8 to 10 weeks.8 The tumors were removed and processed to a single cell suspension with mechanical disruption in the presence of 25 mg collagenase (GIBCO, Burlington, ON) carried out in 100 ml of phosphate-buffered saline (PBS). The polyoma Middle T (PyMidT) cells were then placed in modified Eagle’s medium (MEM) F-11 media containing 10% fetal bovine serum (FBS) (complete F-11) and cultured for 48 hours at 37°C and 5% CO2 before intravenous delivery to syngeneic FVB mice.
Matrigel was purchased from Collaborative Biomedical Products, (Bedford, MA) and was stored at -20°C until use. All cell culture media was purchased from GIBCO (Burlington, ON).
Adenoviruses were propagated in 293 (Ad5 E1-transformed) human embryonic kidney cells maintained in complete F-11 containing 100 µg/ml penicillin and 100 U/ml streptomycin.9 Viruses were purified by double cesium chloride gradient centrifugation following growth in suspension 293 N3S cells using Joklik’s medium and 10% FBS.9 A549 human lung carcinoma cells (ATCC CCL-185) were maintained in MEM media containing 10% FBS.
Human umbilical cord endothelial cells (HUVEC) were obtained from fresh umbilical cords by washing with 2% type 2 collagenase (GIBCO) in Hanks’ buffered saline. HUVECs were maintained in monolayer on culture dishes pre-coated with 2% (w/v) type 2 collagen in PBS, M199 media supplemented with 10 ng/ml endothelial cell growth factor (ICN, Montreal, Quebec) and 10% FBS.10
Recombinant Adenovirus
RNA Extraction and Reverse Transcription-Polymerase Chain Reaction
Total cellular RNA was extracted from quick frozen livers of C57BL/6 mice using Trizol reagent (GIBCO). Reverse transcription reactions were carried out with 0.5, 1.0, 3.0, and 5.0 pg of total RNA in a 20 µl final volume using oligo dT primers supplied with the SuperScript RT reaction kit (GIBCO). Briefly, RNA and oligo dT primers were heated to 70°C for 10 minutes and then placed on ice. 10x RT buffer, MgCl2 dNTPs and DTT were added to the reaction which was subsequently heated to 42°C for 5 minutes after which 1 µl of Superscript reverse transcriptase enzyme was added and the reaction incubated for another 50 minutes. The reaction was stopped by heating to 70°C for 10 minutes and then treated with RNase for 20 minutes at 37°C.
Polymerase chain reactions were carried out in a Perkin Elmer 9600 thermocycler in a final volume of 100 µl using Vent polymerase (NEB, Mississsauga, ON). The angiostatin cDNA is based on the corresponding amino acids 1–32 and 98–458 of the native murine plasminogen11 and the protein sequence of angiostatin derived by enzyme proteolysis.2,12 Two polymerase chain reactions were carried out, one to form the sequence for amino acids 1–32 which also contained the native plasminogen signal sequence and a second reaction which added the bioactive four kringle region of angiostatin. The following primer pairs were used in the first reaction: forward primer, (A) 5'-TGGGGAATTCTTGTTGGCCAGTCCCAACATGGACCATAAGGAAT3', reverse primer (B) 5'-ACC-ACTCGAGAAGAAGCCCCTTGTGTGCTTATGTAGCCAT-CCAT3', which contained an EcoRI and XhoI restriction enzyme site respectively. The second PCR reaction used the following primer pairs: (C) 5'-CTTCTTCTCGAGTGTATCTGTCAGAATGTAAG3', reverse primer, (D) 5'-CCGAAGCTTTTATCATCCTGTCTCTGAGCTCCGCTTCAG3'which contained a stop codon. Primers C and D contained XhoI and HindIII restriction enzyme sites respectively. Cycling conditions for the first section were: denaturation at 94°C, 3 minutes; 94°C, 1 minute; annealing at 55°C, 1 minute; extension at 72°C, 1 minute for 35 cycles. The cycling conditions for the second segment were identical, except that the extension time was increased to 1 minute and 30 seconds. The PCR fragments were run on 1% agarose gels, excised, and isolated using QIAex II gel purification (Qiagen, Chatsworth, CA). The PCR fragments were restriction enzyme digested, re-purified, and ligated into the EcoRI and HindIII polylinker region of the adenovirus shuttle plasmid pACCMV forming a 1170 bp open reading frame which codes for murine angiostatin. The shuttle plasmid pACCMV-angiostatin was co-transfected into 293 cells with pJM17 using calcium phosphate co-precipitation.13 Ad-angiostatin was isolated from a single plaque, expanded in 293 cells and purified by cesium chloride centrifugation.9 The control vector Ad-dl70, without an expressed gene product, was previously described.14 All primers were synthesized at MOBIX (McMaster University, Hamilton, Ontario).
Northern Blots
Total RNA was isolated from A549 cells infected with Ad-dl70 or
Ad-angiostatin at a multiplicity of infection (MOI) of 100 using Trizol
(GIBCO). 20 µg of RNA was loaded per lane and electrophoresed through
a 1% agarose formaldehyde denaturing gel at 20V. The RNA was
transferred overnight by capillary action using 10x saline sodium
citrate to Nytran membranes (ICN). Membranes were hybridized overnight
at 55°C with 
[32P]-CTP random primed
(Pharmacia, Uppsala, Sweden) mouse angiostatin cDNA polymerase chain
reaction product as probe.
Western Blots
Supernatants from HUVEC or A549 cells infected 5 days earlier with Ad-angiostatin or Ad-dl70 at multiplicity of infection (MOI) 100, were incubated overnight at 4°C with 250 µl of lysine-Sepharose (Pharmacia, Uppsala, Sweden) per ml of supernatant. The slurry was collected by centrifugation and washed with 50 mmol/L sodium phosphate buffer, pH 7.4. A 15-µl aliquot was run on a 12% SDS-polyacrylamide gel under non-reducing conditions, and transferred to Immobulon-P membranes (Millipore, Mississauga, ON). The membranes were probed with a 1:500 dilution using a chicken anti-rabbit-plasminogen IgY antibody, followed by a secondary rabbit anti-chicken alkaline phosphatase conjugated antibody (Zymed, San Francisco, CA). A positive control, partially purified rabbit angiostatin derived from human urokinase digested rabbit plasminogen was used as a positive control. Prestained molecular weight markers were purchased from Novagen (Madison, WI).
In Vitro HUVEC Proliferation Assay (Direct Infection and Conditioned Media Assay)
HUVEC cells were plated in 6-well plates pre-coated with 2% gelatin in 3 ml of complete M199 media at a density of 120,000 cells per well. Cells were infected with Ad-dl70, or Ad-angiostatin at a MOI of 100 and were allowed to proliferate for seven days. Two ml of fresh complete M-199 media was added every other day to maintain cell viability. On day 7, the cells were removed with 0.5x trypsin/EDTA (GIBCO), and total cells including those floating in the media were counted using trypan blue (GIBCO) exclusion staining to determine total numbers and viability.
To determined whether the angiostatin gene product was active on endothelial cell proliferation, HUVEC cells were plated in 96-well plates pre-coated with 2% gelatin in complete M199 media at a density of 5000 cells per well overnight. A549 cell culture supernatants infected for 48 hours with Ad-dl70 or Ad-Angiostatin at a MOI of 100 were filter sterilized and UV-irradiated for 30 minutes to inactivate the virus. Fresh M199 media containing 50, 25, or 12.5% conditioned A549 supernatants were added to the HUVEC cells in a final volume of 200 µl. Two days later, the HUVEC cells were labeled with 1 µCi of [3H]thymidine, and incorporation was measured 18 hours later.
In Vivo Matrigel Bioassay for Angiostatin
Matrigel at 4°C was mixed with 100 ng/ml bFGF (GIBCO) alone or in combination with Ad-dl70 (1 x 109 pfu), and Ad-Angiostatin (1 x 109 pfu) in a final volume of 0.5 ml. The Matrigel mixture was then injected subcutaneously into the abdominal midline of C57BL/6 mice, where it polymerized to form a plug. The plug was removed on day 7, fixed in sodium phosphate (0.075 mol/L, pH 7.4)buffered, 10% formalin solution (Fisher Chemicals, Ottawa, ON) for 24 hours and then embedded in paraffin. All tissues were sectioned (5 µm thickness), mounted onto slides and stained with Masson’s trichrome using standard techniques. Total cell invasion was determined by image analysis of the stained sections using a Sony CCD camera and Vidas (Kontron, UK) software. The amount of cell migration was determined by calculating the percent area occupied by the endothelial cells which stained red, versus the total area of the matrigel which stained blue/green.
Electron Microscopy
Portions of paraffin-embedded Matrigel plugs from the bFGF and bFGF plus Ad-angiostatin treatments were extricated. The samples chosen were based on regions of the Masson’s trichrome stained sections that exhibited cellular morphological changes indicating endothelial infiltration, capillary development, or apoptosis.
The Matrigel was removed from the paraffin by heating and then dissolving the sample in xylene. This was followed by re-hydration using 50% ethanol/xylene, 70% ethanol/xylene, 100% ethanol, 90% ethanol, and finally 70% ethanol. The sample was then placed in 0.1 mol/L sodium cacodylate (NaCac) pH 7.4, followed by fixation in 2% glutaraldehyde buffered in 0.1 mol/L NaCac, pH 7.4. All procedures were carried out at room temperature. Treatment with 1% OsO4 in 0.1 mol/L NaCac at 4°C for one hour ensued. The samples were dehydrated using a gradient of ethanol washes ranging from 50% to 100%, followed by propylene oxide treatment. The samples were then embedded in 100% Spurr’s resin, heated overnight at 65°C and sectioned to a thickness of 1 µm. Sections were placed on copper grids and examined using a JEOL 1200 EX Biosystem electron microscope (Tokyo, Japan) at 80 kV.
Lung Gene Expression and Bronchioalveolar Lavage
Ad-Angiostatin or control virus Ad-dl70 were introduced into the lungs of FVB/n mice by intranasal delivery (i.n.). Briefly, 1 x 109 pfu of vector, in a total volume of 30 µl PBS was inserted into the nostrils of anesthetized mice in 15-µl aliquots using a p10 Pipetteman and tip. Seven days post-infection, the lungs were removed, and bronchial alveolar lavage (BAL) fluid was obtained. The anesthetized mice were sacrificed by cutting the abdominal aorta and the entire lung with the attached trachea was removed from the chest cavity. A small diameter (2 mm) section of Tygon tubing attached to a 26-gauge needle and 1 ml syringe was inserted into the lungs through the trachea. BAL fluid was collected from the lungs by twice filling with 150 µl of PBS, which was then aspirated. The lungs were fixed for 24 hours in 10% neutral buffered formalin and paraffin embedded. Sections (5 µm) were hematoxylin and eosin (H&E) stained using standard techniques.
Pulmonary Metastatic Growth of Breast Cancer Cells
3.0 x 105 PyMid-T murine breast adenocarcinoma cells in a volume of 200 µl of PBS were injected as a single cell suspension intravenously into the tail vein of FVB/n mice. This number of cells routinely shows significant metastatic lung nodules within 20 to 25 days after injection. Sixteen days post tumor injection, Ad-angiostatin, and Ad-dl70 (1 x 109 pfu) were delivered intranasally. Lungs were removed on day 24, weighed, and representative lungs from each treatment group were lavaged. Bronchoalveolar lavage fluid 0.5 ml was centrifuged to remove cells and the supernatant was concentrated using 6% trichloroacetic acid precipitation. The recovered protein was resolubilized in 50 µl, 0.5 mol/L Tris buffer, pH 6.8 and run on a 12% SDS-PAGE gel under non-reducing conditions, transferred to Immobulon-P membranes, and Western immunoblotted as described above. The lungs were then formalin-fixed for 24 hours and the total number of surface lung metastases were counted under a stereomicroscope on 6X power. Following paraffin embedding, 5-µm sections were cut and H&E stained to observe tumor vasculature and lung pathology. Statistical significance of differences was tested using Student’s t-test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. No angiostatin mRNA could be found
in the Ad-dl70 infected A549 cells.
|
. Cell culture
supernatants were harvested 5 days after infection and mixed with
lysine-Sepharose to concentrate angiostatin from the media and
subjected to gel electrophoresis. A single band is seen in lane 2 that
has an apparent molecular weight of 50 kd. This single band may
represent two angiostatin isoforms mixed together, as the
unconcentrated culture supernatant (lane 3) gives rise to two distinct
bands between 40 and 50 kd. This doublet may indicate differences in
glycosylation. The chicken anti-rabbit plasminogen IgY antibody
detected rabbit angiostatin (lane 4), derived from the proteolytic
digestion of rabbit plasminogen, as well as bovine plasminogen from the
culture media at 92 kd. However, no bands were detected in the 40 to 50
kd range in the Ad-dl70 control transfected sample (lane 1) indicating
that the bands appearing in the Ad-Angiostatin lanes were not simply
the result of degradation of serum-derived plasminogen. Angiostatin Biological Activity in Vitro
Angiostatin biological activity was determined in vitro
by two methods. HUVEC cells plated to a density of 1.2 x
105
cells per well were directly transduced at
MOI of 100. Control wells were either uninfected, or infected with
control vector Ad-dl70, or Ad-angiostatin. Seven days post-infection,
the cells were gently removed and counted using trypan blue exclusion
staining. Direct transduction of subconfluent HUVEC cells by
Ad-angiostatin led to a 38% decrease in the number of viable cells
seven days after infection. In contrast, uninfected and Ad-dl70
infected cells increased in cell numbers by 276% and 240%,
respectively, during the same period (Figure 1C)
. HUVEC cells, whether
uninfected or infected with control virus proliferated to a confluent
state with the cells exhibiting the normal endothelial phenotype. In
contrast, HUVEC cells infected with Ad-angiostatin were sparse,
rounded, and detached.
To exclude the possibility of a direct viral toxic effect, conditioned
media from A549 cells infected with Ad-angiostatin or Ad-dl70, was
irradiated to inactivate any contaminating virus and was added to
proliferating HUVEC cultures to measure
[3H]thymidine incorporation. Murine angiostatin
was detected by Western immunoblotting in the Ad-angiostatin, infected
A549 conditioned media (Figure 1A
, lane 5). Addition of a range of
dilutions of Ad-Angiostatin conditioned media induced a 54-to 106-fold
inhibition of HUVEC cell proliferation compared to mock-transfected
conditioned media (Figure 1D)
.
Angiostatin Biological Activity in Vivo
Matrigel, an extracellular matrix preparation from Englebreth-Holm-Swarm (EHS) murine sarcoma, is fluid at 4°C but solidifies at 37°C.15 This solid matrix, when combined with angiogenic factors such as bFGF, initiates endothelial cell migration and capillary tube formation into the matrix and is an excellent assay for angiogenesis studies in vivo.15;16
To assess the biological activity of angiostatin derived from the adenoviral vector, mice were injected subcutaneously with Matrigel alone, Matrigel mixed with bFGF (100 ng/ml), or Matrigel plus bFGF plus control virus Ad-dl70 or Ad-angiostatin (1 x 109 pfu).
Histological sections of the Matrigel plugs stained with Masson’s
trichrome are shown in Figure 2
. Seven
days post-infection, the Matrigel alone revealed only a few endothelial
cells invading from the tissue into the edge of the Matrigel in the
absence of any angiogenic factor (Figure 2A)
. In contrast, the addition
of bFGF to the matrigel caused a large increase in endothelial cell
infiltration, and the development of capillary tubes containing red
blood cells is evident (Figure 2B)
. Morphologically, the endothelial
cells appear long and tapered with no distinct nuclear staining.
Addition of the control adenovirus does not affect the process of
endothelial cell invasion or capillary tube formation within the bFGF
impregnated Matrigel as seen in panel C. In sharp contrast, the
Ad-Angiostatin vector inhibited endothelial cell migration into the
center of the matrigel. The cellular invasion was limited to a sharp
line of demarcation near the outer edge of the plug and no vessel
formation is evident (Figure 2, D and E)
. Moreover, the endothelial
cell morphology is distinct, changing from a long tapered phenotype to
a rounded cell with a pyknotic nucleus that exhibits intense nuclear
staining suggesting apoptosis.
|
. Moreover the formation of
functional capillaries containing red blood cells is clearly evident.
In addition, Weibel Palade bodies, the storage granule for von
Willebrand factor (vWf), which are unique to endothelial cells, are
also present. Fibroblasts are not prominent within the matrigel
sections, even though 100 ng/ml of bFGF was added to the Matrigel. A
small number of neutrophils and eosinophils were observed in both
vector treatment groups but there was no sign of an inflammatory
process.
|
.17,18
Granulocytes were not affected
by angiostatin and appeared in a normal, unstimulated state with their
enzyme storage granules intact (not shown). Lung Gene Expression and Bronchoalveolar Lavage
The expression of Ad-angiostatin in the lungs of mice did not
induce any apparent abnormal pathology. Inflammation was limited to the
extent of the control vector and no abnormal pathology was evident at
seven days. Examination of the BAL fluid contents revealed similar
amounts of infiltrating lymphocytes (Table 1)
in both angiostatin and control vector
treated lungs in FVB/n mice. Endothelial cells lining the veins and
arteries appear intact and were not thrombosed. No necrosis or
apoptosis of the lung tissue as a possible result of insufficient blood
perfusion was observed.
|
A model of breast cancer metastasis to the lung was established in FVB/n mice. Mice were injected in the tail vein with 3.0 x 105 PyMidT primary tumor cells and preliminary evidence of nodules on the lung surface appear at 14 to 15 days post injection. On day 16, 1 x 109 pfu control Ad-dl70 or Ad-angiostatin was administered by intranasal delivery. The mice were subsequently sacrificed one week later at which time the lungs were weighed, BAL fluid was collected, and the lung surface metastases counted.
Adenoviral-mediated gene transfer of angiostatin to the lung inhibited
the growth of metastatic tumors as quantified by scoring of surface
metastases. Gross examination of the lungs revealed hard yellowish
tumor nodules on the surface of both treatment groups. However, the
animals in the control treated group had significantly more surface
nodules (98 ± 37; n = 7; Figure 4A
) then those mice which had received
Ad-Angiostatin (18 ± 13; n = 10;
P < 0.0005). The average weight of lungs treated with
control vector alone was 0.72 ± 0.24 g, whereas the
angiostatin-treated animals had a mean lung weight of 0.44 ±
0.05 g (Figure 4B)
. This difference was also highly significant
(P < 0.005). BAL fluid sampled from the lungs
of control treated mice contained 2.3 x 106
(±2.1 x 105) cells/ml which were mainly
macrophages, compared to only 6.4 x 105
(±4.2 x 104) cells/ml in the
angiostatin-treated lungs. This value closely reflects the numbers
obtained from normal non-tumor laden lungs and confirms our earlier
findings that over-expression of angiostatin alone does not appear to
contribute to any inflammation or lung pathology. Western
immunoblotting of the concentrated recovered BAL fluids demonstrated
two distinct bands corresponding to murine angiostatin between 40 and
50 kd in the Ad-angiostatin-treated lungs, but not in the BAL fluid
recovered from control vector-treated lungs (Figure 4C)
. These bands
are consistent with the in vitro culture production of
angiostatin seen in infected HUVEC or A549 cell culture supernatants.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Other reports have used angiostatin derived from enzymatic digestion, or recombinant protein from Pichia pastoris26 and baculovirus.12 Our study has undertaken an alternative approach through an adenoviral gene therapy vector to express the murine angiostatin cDNA and allows for the in situ expression of angiostatin. Recombinant angiostatin has been reported to be fragile and may lose biological activity.27 In addition, recombinant protein may not be correctly glycosylated resulting in a shorter circulating half-life.7 The murine angiostatin cDNA was created by reverse transcription PCR on total RNA isolated from the livers of C57BL/6 mice. Oligonucleotide primers were based on the DNA sequence of murine plasminogen to generate the recombinant murine angiostatin.11 We fused the DNA of the native plasminogen signal sequence, including amino acids 1–32, to the region coding for the four kringle domains at amino acids 98–458. This cDNA was subsequently cloned into the adenovirus shuttle plasmid pACCMV, under the control of the human cytomegalovirus promotor, and rescued with the rescue plasmid pJM-17 to generate Ad-angiostatin.
Angiostatin mRNA was measured in A549 cells 36 hours after infection with Ad-angiostatin. Angiostatin protein production from infected HUVEC and A549 cell culture supernatant was confirmed by Western immunoblotting, using a unique chicken cross-reacting antibody raised against rabbit plasminogen. Concentration of the infected culture supernatants with lysine-Sepharose revealed a band at approximately 50 kd in the Ad-angiostatin, but not in the Ad-dl70-concentrated culture supernatants. No bands were visible in the Ad-dl70 control supernatant lane indicating that protein bands were not an artifact of plasminogen degradation from the cell culture media. The ability of angiostatin to bind to lysine-Sepharose would suggest that the lysine binding sites of the four kringle domains, each containing three intrakringle disulfide bonds, have formed correctly.7,11
Murine angiostatin expressed from the adenovirus vector effectively inhibited the proliferation of HUVEC cells in vitro. This effect was specific to Ad-angiostatin demonstrating that the activity of murine angiostatin is not limited to mouse endothelial cells. HUVECs infected directly with the Ad-angiostatin vector had an overall decrease in viability of 38% as compared to the initial number of viable cells seeded before viral infection. To rule out the possibility of a direct viral-induced toxic effect on endothelial cells, radiation inactivated culture supernatants from Ad-angiostatin infected A549 cells were added to HUVEC cells to measure changes in proliferation. The addition of Ad-angiostatin-infected A549 culture supernatants to subconfluent HUVEC cells demonstrated a significant reduction in HUVEC proliferation as measured by [3H]thymidine incorporation indicating the gene product had a direct inhibitory action on endothelial cells and this inhibition was not due to a toxic effect of the virus on these cells.
We have taken a unique approach to studying the effects of Ad-angiostatin in a Matrigel model of bFGF-induced angiogenesis. Matrigel is an extracellular matrix preparation which is composed primarily of proteoglycan, laminin and collagen.15 We have injected Matrigel subcutaneously alone, or in combination with bFGF and various adenoviral constructs, including Ad-Angiostatin into C57BL/6 mice for one week to study angiogenesis in vivo. Histological staining of the Matrigel sections with Masson’s trichrome revealed no cellular invasion into Matrigel alone. The addition of an angiogenic factor such as bFGF induced a large infiltration of endothelial cells, with the formation of functional capillaries. The inclusion of Ad-angiostatin, but not a control virus, totally eliminated capillary formation and restricted endothelial cell migration to a small region along the edge of the Matrigel. The endothelial cells in the bFGF and bFGF plus control virus-treated sections appear to have normal morphology, whereas the angiostatin-treated cells are rounded and have dark pyknotic nuclei indicative of apoptosis. Angiostatin has been shown to be a potent inducer of apoptosis in endothelial cells.28 Image analysis of the Masson’s trichrome-stained Matrigel sections revealed that the migrating endothelial cells occupied 22% of the Matrigel section for all treatments which included the addition of bFGF, as compared to the Matrigel alone which was not angiogenic. Such an observation may indicate that migrating endothelial cells first respond to the angiogenic stimuli of the recombinant bFGF protein irrespective of the presence of any adenoviral vectors. However, further endothelial cell migration in the Ad-angiostatin-impregnated Matrigel is limited to the interface between the Matrigel and mouse tissues as the angiostatin is synthesized in this region. The angiostatin protein requires additional time to accumulate due to the initial requirement by the cells to take up the vector and synthesize the transgene. Therefore, following the initial angiogenic stimulus by recombinant bFGF, which does not have to synthesized on site, endothelial cells migrate into the Matrigel. However, with time, increasing concentrations of angiostatin limits the migration and induces the apoptosis of the migrating endothelial cells in the Matrigel. Electron microscopy of the Matrigel sections confirmed that the majority of cells within the Matrigel were endothelial cells as noted by the presence of Weibel Palade bodies and the formation of developing capillaries containing red blood cells. Moreover, only the endothelial cells had undergone apoptosis in the angiostatin treated group and no apoptosis was found in the control vector treatment.
Angiostatin-based therapy could be applied to all forms of solid tumors to inhibit the growth of both the primary lesion and any metastatic disease. Anti-angiogenesis therapy may represent a universal treatment for cancer because all solid tumors require neovascularization to grow.
The reduction of a breast cancer metastasis in the lung is an example of how an adenovirus expressing angiostatin could be used to reduce the tumor burden. In the metastatic breast cancer model described here, intranasal delivery reduced the number of tumor nodules in the lungs of treated mice, but did not appear to induce any apparent abnormal pathology to the lung tissue. However, Ad-angiostatin treatment did not totally inhibit the growth of the metastatic nodules, and therefore, angiostatin therapy may need to be combined with traditional therapies like radiation or with gene therapy induced immunotherapy to achieve optimal results.
This report has demonstrated the construction and characterization of an adenoviral vector expressing the angiogenesis inhibitor angiostatin. The biological activity of this gene therapy vector in a Matrigel model not only highlights the biological activity of angiostatin in vivo, but also allows for an unobstructed view of the neovascularization process. The use of such a vector has many applications in treating disease, the foremost being cancer. The specific targeting of tumors as opposed to a systemic administration of recombinant protein would allow for sustained, high concentrations of angiostatin to remain in the tumor environment for maximum efficiency in inhibiting tumor growth.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by the Medical Research Council of Canada (MRC), The Hamilton Health Sciences Corporation, and St. Joseph’s Hospital.
Accepted for publication May 31, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
affect tumorigenicity and response to IL-12 therapy and antiangiogenesis. Immunity 1998, 9:25-34[Medline]
This article has been cited by other articles:
 |
|
 |
|
  V. Schmitz, E. Raskopf, M. A. Gonzalez-Carmona, A. Vogt, C. Rabe, L. Leifeld, M. Kornek, T. Sauerbruch, and W. H Caselmann Plasminogen fragment K1-5 improves survival in a murine hepatocellular carcinoma model Gut, February 1, 2007; 56(2): 271 - 278. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F.-C. Yang, S. Chen, T. Clegg, X. Li, T. Morgan, S. A. Estwick, J. Yuan, W. Khalaf, S. Burgin, J. Travers, et al. Nf1+/- mast cells induce neurofibroma like phenotypes through secreted TGF-{beta} signaling Hum. Mol. Genet., August 15, 2006; 15(16): 2421 - 2437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Peyruchaud, C.-M. Serre, R. NicAmhlaoibh, P. Fournier, and P. Clezardin Angiostatin Inhibits Bone Metastasis Formation in Nude Mice through a Direct Anti-osteoclastic Activity J. Biol. Chem., November 14, 2003; 278(46): 45826 - 45832. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Dabrosin, S. Gyorffy, P. Margetts, C. Ross, and J. Gauldie Therapeutic Effect of Angiostatin Gene Transfer in a Murine Model of Endometriosis Am. J. Pathol., September 1, 2002; 161(3): 909 - 918. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Margetts, S. Gyorffy, M. Kolb, L. Yu, C. M. Hoff, C. J. Holmes, and J. Gauldie Antiangiogenic and Antifibrotic Gene Therapy in a Chronic Infusion Model of Peritoneal Dialysis in Rats J. Am. Soc. Nephrol., March 1, 2002; 13(3): 721 - 728. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
