help button home button Am J Pathol Angiogenesis Meeting
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Roth, J. M.
Right arrow Articles by Brooks, P. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Roth, J. M.
Right arrow Articles by Brooks, P. C.
(American Journal of Pathology. 2005;166:901-911.)
© 2005 American Society for Investigative Pathology

Recombinant {alpha}2(IV)NC1 Domain Inhibits Tumor Cell-Extracellular Matrix Interactions, Induces Cellular Senescence, and Inhibits Tumor Growth in Vivo

Jennifer M. Roth, Abebe Akalu, Anat Zelmanovich, Desiree Policarpio, Bruce Ng, Shannon MacDonald, Silvia Formenti, Leonard Liebes and Peter C. Brooks

From the Departments of Radiation Oncology and Cell Biology, New York University School of Medicine, New York, New York


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Cellular interaction with the extracellular matrix is thought to be a critical event in controlling angiogenesis and tumor growth. In our previous studies, genetically distinct noncollagenous (NC) domains of type-IV collagen were shown to interact with integrin receptors expressed on the surface of endothelial cells. Moreover, these NC1 domains were shown to inhibit angiogenesis in vivo. Here, we provide evidence that a recombinant form of the {alpha}2(IV)NC1 domain of type-IV collagen could bind integrins {alpha}1ß1 and {alpha}vß3 expressed on melanoma cells and inhibit tumor cell adhesion in a ligand-specific manner. Systemic administration of recombinant {alpha}2(IV)NC1 domain potently inhibited M21 melanoma tumor growth within full thickness human skin and exhibited a dose-dependent inhibition of tumor growth in nude mice. Interestingly, {alpha}2(IV)NC1 domain enhanced cellular senescence in tumor cells in vitro and in vivo. Taken together, these results suggest that recombinant {alpha}2(IV)NC1 domain is not only a potent anti-angiogenic reagent, but it also directly impacts tumor cell behavior. Thus, {alpha}2(IV)NC1 domain represents a potent inhibitor of tumor growth by impacting both endothelial and tumor cell compartments.

In recent years a great deal of attention has been focused on understanding the molecular mechanisms by which the extracellular matrix (ECM) regulates angiogenesis and tumor growth.1-6 In this regard, the basement membrane is a specialized form of ECM composed of a complex network of collagenous and noncollagenous glycoproteins and proteoglycans. This specialized compartment of the ECM lines epithelial sheets, surrounds blood vessels and nerves, and is composed predominately of col-lagen type-IV, laminin, and nidogen/entactin. Numerous studies have demonstrated that while basement membranes can provide mechanical support for cells and tissues, it also regulates a great diversity of other cellular processes including adhesion, migration, invasion, differentiation, cell cycle, and gene expression. Importantly, all these events have been implicated in the regulation of angiogenesis and tumor growth and are currently under intensive study in an attempt to gain a more complete understanding of the molecular mechanisms contributing to normal physiology and human disease. In fact, recent studies have identified several endogenous inhibitors of angiogenesis and tumor growth, which are derived from specific domains of ECM components such as collagen.1-6

Collagen type-IV is a major component of the basement membrane and to date at least six genetically distinct collagen type-IV chains [{alpha}1(IV) to {alpha}6(IV)] have been identified and characterized.7,8 One of the most widely distributed forms of collagen type-IV is composed of two {alpha}1(IV) chains and one {alpha}2(IV) chain organized in a triple helical manner. Collagen type-IV is organized into a number of functional domains including an N-terminal 7S domain, a central triple helical region, and a C-terminal noncollagenous (NC1) domain. Previous studies have identified unique functions for these structural domains, including the regulation of chain selection during the formation of the triple helical molecule, network formation, and regulation of integrin-dependent cell adhesion and migration.1-8 In fact, in studies using Hydra vulgaris, exogenous addition of purified NC1 domains of collagen type-IV resulted in disruption of ECM formation and inhibition of morphogenesis.9 These important early findings led to studies that suggested that recombinant NC1 domains disrupt angiogenesis.10 Importantly, the recent elucidation of the crystal structure of an NC1 domain hexamer will likely provided critical insight into the molecular mechanisms regulating collagen network assembly as well as contribute to a better understanding of the mechanisms by which isolated NC1 domains impact both physiological and pathological processes.11

Cellular interactions with collagen type-IV are facilitated predominately by integrin cell-surface receptors.12,13 Integrins are a diverse group of transmembrane receptors composed of noncovalently associated {alpha} and ß chains that combine to give heterodimers with distinct functional activities. Integrins play critical roles in facilitating bidirectional signal transduction between the local ECM and cells. In addition, studies have indicated that a number of ß1-containing integrins can facilitate cellular interactions with triple helical collagen type-IV including {alpha}1ß1, {alpha}2ß1, and {alpha}3ß1.12,13 Moreover, evidence has also been provided that specific NC domains can interact with distinct integrins including, {alpha}1ß1, {alpha}3ß1, {alpha}5ß1 {alpha}vß3, and {alpha}vß5.14-19 These findings suggest that NC domains may exhibit specificity as to which integrin receptors they interact with, which in turn, may contribute to the regulation of distinct cellular events.

Recent studies from many laboratories have identified unique functions and biological activities for collagen NC1 domains.4,7,14 In fact, investigators have demonstrated that recombinant forms of collagen type-IV NC1 domains including {alpha}1(IV) (arrestin), {alpha}2(IV) (canstatin), {alpha}3(IV) (tumstatin), and the {alpha}6(IV) can potently inhibit angiogenesis and tumor growth in vivo.14-19 In contrast, little if any anti-angiogenic or anti-tumor activity was observed with similar NC1 domains derived from the {alpha}4 and {alpha}5 chains of collagen type-IV.14 Although significant progress has been made in understanding the molecular mechanisms by which {alpha}3(IV) NC1 domain (tumstatin) inhibits angiogenesis and tumor growth, less is known concerning the anti-angiogenic and anti-tumor effects of {alpha}2(IV)NC1 domain (canstatin). To this end, our previous studies suggested that {alpha}2(IV)NC1 domain could interact with integrins {alpha}3ß1, {alpha}vß3, and {alpha}vß5 on the surface of endothelial cells and potently inhibit growth factor-induced angiogenesis and tumor growth in the chick embryo.14

To expand these studies, we examined whether recombinant {alpha}2(IV)NC1 domain could specifically interact with tumor cells and furthermore, whether it could directly effect tumor cell behavior. Here, we provide evidence that human tumor cells such as M21 melanoma and SKOV-3 ovarian carcinoma cells can directly interact with {alpha}2(IV)NC1 domain in an integrin-specific manner that is distinct from that of endothelial cells. In addition, our results suggest for the first time that recombinant {alpha}2(IV)NC1 domain can specifically inhibit tumor cell adhesion to collagen type-IV and induce tumor cell senescence in a ligand-dependent manner. Finally, we demonstrate that {alpha}2(IV)NC1 domain can potently inhibit tumor angiogenesis and tumor growth in multiple models. Taken together, our findings suggest that {alpha}2(IV)NC1 domain is not only an anti-angiogenic molecule but can impact both the endothelial and tumor cell compartments and that its administration may represent an important therapeutic approach for the treatment of human malignancies.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
Chemicals, Reagents, and Antibodies

Purified ECM proteins including collagen (pepsin solubilized) types-I and -IV lacking 7S and NC1 domains, fibronectin, and vitronectin were obtained from Sigma (St. Louis, MO). Bovine serum albumin (BSA), ethanol, methanol, and acetone were all obtained from Sigma. OTC embedding compound was obtained from VWR International (Bridgeport, NJ). Monoclonal antibodies (mAbs) LM609 (anti-{alpha}vß3), P4C10 (anti-ß1), and P1F6 (anti-{alpha}vß5) were kindly provided by Dr. David Cheresh, Scripps Research Institute (La Jolla, CA). mAbs directed to {alpha}1, {alpha}2, and {alpha}3 integrin subunits were obtained from Chemicon (Temecula, CA). Fluorescein isothiocyanate- and rhodamine-labeled goat anti-mouse secondary antibodies were obtained from BioGenix Inc. (San Ramos, CA). Polyclonal antibody AB769 directed to CD-31 was obtained from BD Pharmingen (San Diego, CA). X-Gal was obtained from Promega (Madison, WI). ApopTag apoptosis detection kit was obtained from Roche (Indianapolis, IN). Recombinant {alpha}2(IV)NC1 domain was kindly provided by BioStratum Inc. (Durham, NC). Recombinant {alpha}5(IV)NC1 domain was kindly provided by Dr. Billy Hudson (Vanderbilt, TN). The control peptide used was a commercially synthesized 25-mer based on the sequence from human {alpha}2(IV) chain of collagen type-IV. The specific amino acid sequence of the control peptide is NH2-CKGIDMPGTPGLKGDRGSPGMDGFC-COOH.

Cells and Cell Culture

Human M21 melanoma cell line was kindly provided by Dr. David Cheresh (Scripps Research Institute). Human SKOV-3 ovarian carcinoma cells were obtained from American Type Culture Collection (Rockville, MD). Cell lines were maintained in Dulbecco’s modified Eagle’s medium (Life Technologies, Inc., Grand Island, NY) supplemented with 10% fetal bovine serum (Hyclone, Logan, UT), 1.0% sodium pyruvate, glutamate, and Pen-Strep (Life Technologies, Inc.). Cells were maintained as subconfluent cultures before use and harvested with trypsin-ethylenediaminetetraacetic acid (Life Technologies, Inc.).

Cell Adhesion Assay

Cell adhesion assays were performed essentially as described.20 Briefly, nontissue culture 48-well plates were coated with either recombinant {alpha}2(IV)NC1 domain (100 µg/ml) or purified ECM proteins including collagen-I and -IV (10 µg/ml) fibronectin (10 µg/ml), or vitronectin (5 µg/ml) for 18 hours at 4°C. Subconfluent cells (M21 and SKOV-3) were harvested, washed, and resuspended in adhesion buffer (RPMI 1640 medium containing 1.0 mmol/L MgCl2, 0.2 mmol/L MnCl2, and 0.5% BSA) in the presence or absence of mAbs (75 µg/ml) or recombinant {alpha}2(IV)NC1 and {alpha}5(IV)NC1 domains (100 µg/ml). Cells were added to the coated wells in a total volume of 200 µl and allowed to attach for 30 to 60 minutes at 37°C. Nonattached cells were removed by washing and attached cells were stained with crystal violet.20 Wells were washed and cell-associated crystal violet was eluted with 100 µl of 10% acetic acid. Cell adhesion was quantified by measuring the optical density of the eluted crystal violet at a wavelength of 600 nm.20

Chick Embryo Tumor Growth Assay

The chick embryo tumor growth assays were performed essentially as described.21 Briefly, 10-day-old chick embryos were prepared by separating the chorioallantoic membrane (CAM) from the shell membrane. Single cell suspensions from subconfluent cultures of SKOV-3 cells (5.0 x 106/embryo) were added to the CAM in a total volume of 40 µl. Twenty-four hours later, the embryos were either untreated or injected intravenously with a single dose of recombinant {alpha}2(IV)NC1 or {alpha}5(IV)NC1 domains (30 µg/embryo) in a total volume of 100 µl per embryo. The embryos were allowed to grow for a total of 7 days at which time they were harvested and the tumors were resected, trimmed free of surrounding CAM tissue, washed, and wet weights determined. Experiments were performed twice with 5 to 10 embryos per condition.

Nude Mouse Tumor Growth Assay

Subconfluent M21 melanoma cells were harvested, washed, and resuspended in sterile phosphate-buffered saline (PBS). Tumor cells (2.0 x 106) were injected subcutaneously into nude mice.22 Three days after tumor cell injections, when small palpable tumors were detected, mice were either untreated or injected (intraperitoneally) with recombinant {alpha}2(IV)NC1 domain or control normal mouse IgG at various concentrations. The mice were treated daily for a total of 21 days. To assess the effects of {alpha}2(IV)NC1 domain on pre-existing tumors, tumors were allowed to grow to ~80 mm3 (day 6) before treatment was initiated. Tumor size was monitored with calipers and tumor volumes were estimated using the formula V = L2 x W/2, where V = volume, L = length, and W = width. Experiments were completed two to three times with similar results.

Human/Mouse Chimeric Model

The human/mouse chimeric tumor growth model was performed essentially as described with some modifications.23 Briefly, fresh human neonatal foreskins (Cooperative Human Tissue Network, Cleveland, OH) were surgically transplanted on nude mice.23 Human M21 melanoma cells were injected within the human skin graft.23 Tumors were allowed to grow within the human skin for 3 days until small palpable tumors were detected. Three days after intradermal injection of tumor cells, mice were treated with daily intraperitoneal injections of {alpha}2(IV)NC1 domain or control for a total of 26 days. Tumor size was monitored with calipers and tumor volumes were estimated using the formula V = L2 x W/2, where V = volume, L = length, and W = width. Experiments were completed two times with similar results.

Immunohistochemical and Immunofluorescence Analysis

Tumors derived from control and {alpha}2(IV)NC1 domain-treated mice were dissected, washed, embedded in OTC, snap-frozen, and 4.0-µm sections were cut with a cryostat.22 To quantify tumor-associated angiogenesis, sections were incubated with 2.5% BSA in PBS to block any nonspecific binding. Endogenous peroxidase activity was quenched by incubation with hydrogen peroxide (0.03%). Tissues were next incubated with anti-CD-31 antibody at a concentration of 10 µg/ml for 2 hours at 37°C, washed with PBS, and incubated with peroxidase-labeled goat anti-rabbit secondary antibody. Quantification of CD-31-positive blood vessels was performed essentially as described24 with some modifications. The number of CD-31-positive blood vessels per microscopic field was counted in hot spots (fields containing at least five vessels). Ten fields were counted for each tumor section with three tumors per condition.

To evaluate the relative levels of cellular apoptosis, we used the ApopTag apoptosis detection kit. Briefly, tumors from control and {alpha}2(IV)NC1 domain-treated mice were dissected, washed, embedded in OTC, snap-frozen, and 4.0-µm sections were cut with a cryostat.22 To quantify cellular apoptosis, sections were incubated with 2.5% BSA in PBS to block any nonspecific binding. Tumor sections were stained with ApopTag apoptosis detection reagents according to the manufacturer’s instructions. The relative levels of apoptosis were estimated by laser-scanning image analysis.25 Briefly, stained tissue sections were scanned using a Kodak ID imaging system (Eastman-Kodak, Rochester, NY) and the pixel density of positive staining regions (x200 microscopic fields) n = 3 per tumor specimen was quantified using Kodak image analysis software as has been described previously.25 In further experiments, co-staining was performed to determine whether CD-31-positive blood vessels were undergoing apoptosis. Briefly, tumors from control and {alpha}2(IV)NC1 domain-treated mice were dissected, washed, embedded in OTC, snap-frozen, and 4.0-µm sections were cut with a cryostat. The tissues were next co-incubated with both ApopTag reagents and anti-CD-31 antibody as we have previously described.26 Sections were washed and incubated with rhodamine-conjugated goat anti-rabbit secondary as described above. Tumor sections were analyzed and photographed with a BX20 Olympus compound microscope fitted with epifluorescences (New Jersey Scientific) with an attached charge-coupled device digital camera.

Analysis of Cellular Senescence

To examine the levels of cellular senescence within tumors, we assessed the relative levels of senescence-associated ß-galactosidase (SA-ß-Gal) as previously described.27,28 Briefly, tumors from control and {alpha}2(IV)NC1 domain-treated mice were dissected, washed, embedded in OTC, snap-frozen, and 4.0-µm sections were cut with a cryostat. To quantitate cellular senescence, sections were first fixed in 3% formaldehyde for 2 to 3 minutes then washed with PBS four times for 5 minutes each. The tissue sections were stained with SA-ß-Gal staining solution containing 1.0 mg/ml X-Gal in 20 mmol/L sodium phosphate, 20 mmol/L citric acid, 5 mmol/L potassium ferricyanide, 2 mmol/L magnesium hexacyano, and 150 mmol/L sodium chloride at pH 6.0.27,28 The tissues were incubated for 24 to 48 hours at 37°C, washed three times in H2O, and mounted with Permount (Fisher, Pittsburgh, PA). Tissue sections were analyzed and photographed with a BX20 Olympus compound microscope fitted with a charge-coupled device digital camera. Quantification of the relative levels of senescence was performed by scoring (+, ++, +++) the tissue sections for cellular expression of SA-ß-Gal. In particular, five x200 microscopic fields were evaluated for each of three distinct tumors per condition. Tissues with an average of less than two positive cells per field were given a score of +, two to five positive cells per field ++, and greater than five positive cells per field +++.

To assess the relative levels of senescence in cells grown in vitro, a similar staining procedure was applied. Briefly, 48-well nontissue culture plates were coated with either collagen type-IV or other ECM components at a concentration of 10 µg/ml. Subconfluent cells were harvested, washed, and resuspended in 1% serum containing medium and plated in the presence or absence of {alpha}2(IV)NC1 domain or controls. The cells were allowed to incubate for 12 to 24 hours at 37°C. The cells were next carefully washed two times with PBS and fixed with 2.0% formaldehyde and 0.2% glutaraldehyde for 2 minutes at room temperature. The cells were washed three times with PBS and stained with SA-ß-Gal staining solution as described above for 12 to 24 hours.27,28 Tumor cells were analyzed and photographed with a BX20 Olympus compound microscope fitted with a charge-coupled device digital camera. In vitro tumor cell senescence was quantified by counting the number of enlarged granular cells staining positive for SA-ß-Gal per x200 microscopic field. Three fields were evaluated for each well with three wells per condition.

Statistical Analysis

Statistical analysis was performed using the InStat statistical program for Macintosh computers. Data were analyzed for statistical significance using Student’s t-test. P values <0.05 were considered significant.


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
Immobilized Recombinant {alpha}2(IV)NC1 Domain Supports Tumor Cell Adhesion in Vitro

ECM components such as collagen type-IV are known to support integrin-dependent tumor cell interactions. Our previous studies, as well as studies from other laboratories have demonstrated that genetically distinct NC domains can interact with endothelial cells and support cell adhesion in vitro.14,29 To expand these studies, we assessed the capacity of recombinant {alpha}2(IV)NC1 domain to directly support tumor cell adhesion. Nontissue culture-treated plates were coated with either recombinant {alpha}2(IV)NC1 domain or BSA as a control. Tumor cells (human M21 melanoma or SKOV-3 ovarian carcinoma) were allowed to adhere to the coated substrates. As shown in Figure 1 , both human M21 melanoma and SKOV-3 ovarian carcinoma cells readily attached to immobilized {alpha}2(IV)NC1 domain while exhibiting little interactions with BSA. These findings are in agreement with our previous studies with endothelial cells and suggest that recombinant {alpha}2(IV)NC1 domain of collagen type-IV can specifically interact with these human melanoma and carcinoma cells.



View larger version (20K):
[in this window]
[in a new window]
  		Figure 1. Immobilized {alpha}2(IV)NC1 domain supports tumor cell adhesion. Nontissue culture-treated 48-well plates were coated with either recombinant {alpha}2(IV)NC1 domain or BSA (100 µg/ml). Subconfluent tumor cells (M21 and SKOV-3) were harvested, washed, and resuspended in adhesion buffer and allowed to attach to the coated wells. Cell adhesion was quantified by measuring the optical density of eluted dye from adherent cells at a wavelength of 600 nm. Data bars represent the mean optical density ± SD from triplicate wells. Experiments were completed three times with similar results.

 
Tumor Cell Interactions with Recombinant {alpha}2(IV)NC1 Domain Depend on Integrin Receptors

Integrin receptors are considered the major family of cell surface receptors mediating cell-ECM interactions.13 Previous studies have suggested that integrin receptors expressed on endothelial cells can interact with distinct NC1 domains of collagen type-IV.14-19 In fact, our studies indicated that endothelial cells interact with immobilized {alpha}2(IV)NC1 domain by multiple integrins, including {alpha}vß3, {alpha}vß5, and {alpha}3ß1.14 To assess whether melanoma cells interact with {alpha}2(IV)NC1 domain using similar sets of integrins, in vitro cell adhesion assays were performed in the presence or absence of function blocking anti-integrin antibodies. As shown in Figure 2A , a monoclonal antibody (mAb) specifically directed to the ß1 integrin subunit inhibited M21 cell adhesion to {alpha}2(IV)NC1 domain by ~60%, whereas a function-blocking antibody directed to {alpha}vß5 had little if any effect. Interestingly, a mAb directed to integrin {alpha}vß3 blocked M21 cell adhesion to {alpha}2(IV)NC1 domain by ~40%, suggesting a role for both {alpha}vß3 and at least one ß1-containing integrin. Importantly, although our previous studies suggested that endothelial cells can interact with {alpha}2(IV)NC1 domain using {alpha}vß3, {alpha}vß5, and {alpha}3ß1, M21 melanoma cells do not use {alpha}vß5, but rather {alpha}vß3 and at least one other ß1-containing integrin.



View larger version (21K):
[in this window]
[in a new window]
  		Figure 2. Tumor cell adhesion to immobilized {alpha}2(IV)NC1 domain is integrin-dependent. Nontissue culture-treated 48-well plates were coated with recombinant {alpha}2(IV)NC1 domain (A and B) or various purified ECM proteins (C–G). Subconfluent M21 melanoma and SKOV-3 carcinoma cells were resuspended in adhesion buffer in the presence or absence of specific antagonists and allowed to attach to the coated wells. Cell adhesion was quantified by measuring the optical density of eluted dye from adherent cells at a wavelength of 600 nm. Data bars represent the mean optical density ± SD from triplicate wells. A and B: Effects of anti-integrin mAbs (75 µg/ml) on M21 melanoma cell attachment to immobilized {alpha}2(IV)NC1 domain. C–G: Effects of soluble recombinant {alpha}2(IV)NC1 domain (100 µg/ml) on M21 (C–F) and SKOV-3 cell (G) attachment to immobilized ECM proteins. NT, no treatment; anti-ß1, mAb P4C10; anti-{alpha}vß5, mAb P1F6; anti-{alpha}vß3, mAb LM609; anti-{alpha}1, mAb 1973Z; anti-{alpha}2, mAb 1950Z; anti-{alpha}3, mAb 1952Z; {alpha}5(IV)NC1, recombinant NC1 domain from the {alpha}5 chain of human collagen type-IV. Experiments were completed two to three times with similar results.

 
To determine which ß1 integrin may contribute to M21 melanoma cell interactions with {alpha}2(IV)NC1 domain, similar cell adhesion assays were performed in the presence or absence of function-blocking antibodies directed to distinct collagen-binding integrin {alpha} subunits. As shown in Figure 2B , a function-blocking antibody directed to the {alpha}1 subunit inhibited cell adhesion by ~50%, whereas antibodies directed to the {alpha}2 and {alpha}3 integrin subunits exhibited little if any effect. Moreover, a combination of anti-{alpha}1 and anti-{alpha}vß3 mAbs inhibited M21 cell adhesion by ~85% as compared to controls (data not shown). These data suggest that M21 melanoma cells predominately use {alpha}1ß1 and {alpha}vß3 integrins to facilitate tumor cell adhesion to {alpha}2(IV)NC1 domain. Given these results, it is possible that addition of soluble recombinant {alpha}2(IV)NC1 domain may inhibit tumor cell adhesion to specific ECM components. To test this possibility, we allowed M21 melanoma cells to attach to purified ECM proteins in the presence or absence of soluble {alpha}2(IV)NC1 domain. As shown in Figure 2C , soluble {alpha}2(IV)NC1 domain (100 µg/ml) inhibited M21 cell adhesion to collagen type-IV by ~50% as compared to control {alpha}5(IV)NC1 domain. In contrast, {alpha}2(IV)NC1 domain failed to inhibit M21 cell adhesion to other ECM proteins including collagen type-I, fibronectin, and vitronectin (Figure 2; D to F) . Importantly, soluble {alpha}2(IV)NC1 domain did not detach cells that had already bound to these ECM ligands (data not shown). To confirm the anti-adhesive activity of recombinant {alpha}2(IV)NC1 domain in a second independent tumor cell type, we examined human SKOV-3 ovarian carcinoma cells. As shown in Figure 2G , {alpha}2(IV)NC1 domain inhibited SKOV-3 cell adhesion to collagen type-IV by ~50% as compared to controls. Consistent with our studies on M21 melanoma cells, SKOV-3 carcinoma cells also primarily use {alpha}1ß1 and {alpha}vß3 to interact with immobilized {alpha}2(IV)NC1 domain (data not shown). These results indicate that the inhibitory activity of {alpha}2(IV)NC1 domain may be associated with an ECM- and/or integrin-specific mechanism.

Recombinant {alpha}2(IV)NC1 Domain Inhibits SKOV-3 Tumor Growth in Vivo

Tumor cell interactions with the ECM have been shown to play a critical role in regulating tumor growth and progression.30-34 Therefore we examined the effects of recombinant {alpha}2(IV)NC1 domain on SKOV-3 tumor growth in vivo using the chick embryo tumor growth model. SKOV-3 carcinoma cells were implanted on the CAMs of 10-day-old chick embryos. Twenty-four hours later the embryos were treated with a single intravenous injection of {alpha}2(IV)NC1 domain or control {alpha}5(IV)NC1 domain. At the end of a 7-day incubation period, the resulting tumors were removed and wet weights determined. As shown in Figure 3 , a single injection of recombinant {alpha}2(IV)NC1 domain significantly (P = 0.035) inhibited SKOV-3 ovarian tumor growth by ~50% as compared to either no treatment or control. Taken together, these finding suggest that the anti-adhesive and anti-tumor effects of {alpha}2(IV)NC1 domain are not restricted to M21 melanoma and may impact a variety of tumors from distinct histological origins.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 3. Recombinant {alpha}2(IV)NC1 domain inhibits SKOV-3 carcinoma tumor growth in vivo. Subconfluent SKOV-3 carcinoma cells were implanted on the CAMs of 10-day-old chick embryos. Twenty-four hours later the embryos were treated with a single intravenous injection of {alpha}2(IV)NC1 or {alpha}5(IV)NC1 domains (30 µg/embryo). At the end of a 7-day incubation period the resulting tumors were removed and wet weights determined. Data bars represent the mean tumor weights ± SE from 8 to 10 embryos per condition. Experiments were completed two times with similar results.

 
Recombinant {alpha}2(IV)NC1 Domain Inhibits Tumor Growth in Murine Models

Our studies as well as work from other laboratories have provided evidence that specific NC1 domains from collagen type-IV can inhibit growth factor- and tumor-induced angiogenesis in vivo.14-19 Moreover, studies have indicated that the growth of malignant tumors depends on angiogenesis. In this regard, we expanded our previous studies to examine the effects of recombinant {alpha}2(IV)NC1 domain on tumor growth in murine models. To determine whether recombinant {alpha}2(IV)NC1 domain exhibited a dose-dependent inhibition of tumor growth, human M21 melanoma cells were injected subcutaneously in nude mice. After development of palpable tumors (3 days), mice were treated daily (intraperitoneally) with {alpha}2(IV)NC1 domain at various concentrations. As shown in Figure 4A , administration of recombinant {alpha}2(IV)NC1 domain resulted in a dose-dependent inhibition of tumor growth with maximal inhibition of ~70% (P < 0.05) observed at a dose of 2.5 mg/kg. Administration of control protein at doses up to 5.0 mg/kg had little if any effect. In similar experiments, M21 tumors were allowed to grow until they reached a mean tumor size of ~80 mm3 (6 days) at which time, mice were treated daily with {alpha}2(IV)NC1 domain at a concentration of 2.5 mg/kg. As shown in Figure 4B , {alpha}2(IV)NC1 domain significantly (P < 0.05) inhibited the growth of established M21 melanoma tumors by ~60% as compared to controls.



View larger version (28K):
[in this window]
[in a new window]
  		Figure 4. Recombinant {alpha}2(IV)NC1 domain inhibits tumor growth in vivo. The effects of recombinant {alpha}2(IV)NC1 domain on the growth of human M21 melanomas were examined in murine xenograft models. A: Tumor-bearing mice were treated (intraperitoneally) daily with increasing concentrations of {alpha}2(IV)NC1 domain or control (normal mouse IgG). B: Tumors (M21) were allowed to grow to a mean size of ~80 mm3 before daily treatments (intraperitoneally) with {alpha}2(IV)NC1 domain or control (2.5 mg/kg). Data bars represent the mean tumor volumes ± SE from five to six mice per condition. C: Human M21 melanoma cells were injected into full-thickness human skin grafts transplanted on the backs of nude mice. After establishment of tumors (3 days), mice were treated (intraperitoneally) daily with {alpha}2(IV)NC1 domain (2.5 mg/kg) or vehicle control (PBS). Data bars represent the mean tumor volumes ± SE from five to six mice per condition.

 
A common limitation of many murine xenograft models involves the growth of malignant tumors in tissue microenvironments not typically observed in the human disease. To confirm the effects of {alpha}2(IV)NC1 domain on melanoma tumor growth, we used the human/mouse chimeric model in which human M21 melanoma cells were allowed to grow within full thickness human skin transplanted on nude mice. In agreement with our previous findings, daily administration of {alpha}2(IV)NC1 domain in mice harboring M21 tumors growing within full thickness human skin resulted in significant (P < 0.05) inhibition by ~75% (Figure 4C) . These findings suggest that recombinant {alpha}2(IV)NC1 domain not only inhibits human M21 tumor growth within the subcutaneous compartment, but also blocks tumor growth within full thickness human skin where human cutaneous melanomas typically arise.

Recombinant {alpha}2(IV)NC1 Domain Inhibits Tumor-Associated Angiogenesis in Vivo

Previously published reports have indicated that {alpha}2(IV)NC1 domain can inhibit angiogenesis stimulated by purified growth factors and certain tumors.14,35-37 To assess the effects of recombinant {alpha}2(IV)NC1 domain on tumor-associated angiogenesis in our models, we examined M21 tumors from either control or {alpha}2(IV)NC1 domain-treated mice. Frozen sections from these tumors were stained for the presence of CD-31, a known marker of blood vessels. Tumors from mice treated with {alpha}2(IV)NC1 domain exhibited a marked reduction in tumor-associated blood vessels from areas near the margin of the tumors. To quantify the apparent change in microvascular density, tumor-associated angiogenesis was estimated by counting the number of blood vessels from 10 microscopic fields from each of three distinct tumors per experimental condition.24 As shown in Figure 5 , tumor blood vessels from mice treated with {alpha}2(IV)NC1 domain were significantly (P = 0.001) reduced by ~45% as compared to controls, confirming the anti-angiogenic activity of {alpha}2(IV)NC1 domain. These findings suggest that the inhibition of tumor growth observed may be associated at least in part with inhibition of tumor angiogenesis.



View larger version (18K):
[in this window]
[in a new window]
  		Figure 5. Recombinant {alpha}2(IV)NC1 domain inhibits tumor-associated angiogenesis in vivo. Tumors (M21) from control and {alpha}2(IV)NC1-treated mice were snap-frozen and 4.0-µm tissue sections prepared. Tumor blood vessels were examined by immunohistochemistry using anti-CD-31 antibody. CD-31-positive vessel counts were performed on 10 microscopic fields from three tumors for each experimental condition. Data bars represents the mean CD-31 vessel counts ± SE from the three tumors per experimental condition. Original magnifications, x200.

 
Elevated Levels of Apoptosis in M21 Tumors from Mice Treated with Recombinant {alpha}2(IV)NC1 Domain

Previous studies have suggested that {alpha}2(IV)NC1 domain can inhibit endothelial cell proliferation and induce endothelial but not tumor cell apoptosis in vitro.35-37 To examine potential mechanisms by which {alpha}2(IV)NC1 domain may inhibit tumor growth, we examined M21 tumors from control and {alpha}2(IV)NC1 domain-treated mice for the relative levels of apoptosis by terminal dUTP nick-end labeling (TUNEL) staining. Frozen sections of tumors from control and {alpha}2(IV)NC1 domain-treated mice were analyzed using the ApopTag apoptosis detection reagent. In control-treated tumors, random apoptotic staining was detected in numerous scattered cells staining TUNEL-positive (Figure 6A) . In contrast, elevated levels of apoptosis were detected localized in discrete foci (Figure 6A , middle and right) throughout tumors treated with {alpha}2(IV)NC1 domain (Figure 6A) . In fact, tumors from mice treated with {alpha}2(IV)NC1 domain (2.5 mg/kg) were associated with a significant (P = 0.043) increase in apoptosis by approximately fourfold as compared to controls (Figure 6B) . Importantly, {alpha}2(IV)NC1 domain failed to induce apoptosis in tumor cells in vitro (data not shown) suggesting that the apoptosis observed within the tumors may be because of indirect mechanisms. Given previous reports that {alpha}2(IV)NC1 domain can induce apoptosis in endothelial cells but not in tumor cells, we examined whether angiogenic blood vessels within the tumor may be undergoing apoptosis. To this end, we co-stained tumor sections from control and {alpha}2(IV)NC1 domain-treated mice with an antibody directed to CD-31 to mark blood vessels and with ApopTag reagent to detect apoptotic cells. As shown in Figure 6C (top left) CD-31-positive (red) tumor blood vessels within control-treated tumors could be readily detected. Moreover, little if any apoptosis (green) was observed in association with tumor blood vessels. In contrast, apoptosis (green) could be detected in association with (yellow co-localization) blood vessels from {alpha}2(IV)NC1 domain-treated tumors (Figure 6C , top right). It is important to point out that although examples of apoptotic blood vessels could be detected (yellow), not all vessels were apoptotic. Interestingly, tumors from mice treated with {alpha}2(IV)NC1 domain were also associated with elevated levels of apoptosis in cells other than CD-31-positive endothelial cells (Figure 6C , bottom). Taken together, these results suggest that {alpha}2(IV)NC1 domain treatment may cause enhanced apoptosis within both nonendothelial cells and tumor-associated blood vessels in vivo.



View larger version (68K):
[in this window]
[in a new window]
  		Figure 6. Tumors from mice treated with recombinant {alpha}2(IV)NC1 domain exhibit enhanced apoptosis. Tumors (M21) from control and {alpha}2(IV)NC1 domain-treated mice were snap-frozen and 4.0-µm tissue sections prepared. Tumor-associated apoptosis was detected by TUNEL staining and tumor blood vessels were examined by immunohistochemistry using anti-CD-31 antibody. A: Representative examples of tumors exhibiting apoptosis within control and {alpha}2(IV)NC1 domain-treated M21 melanoma tumors after TUNEL staining (brown). B: Quantification of the relative levels of apoptosis within tumors from control and {alpha}2(IV)NC1 domain-treated mice. Data bars represent the mean apoptotic staining ± SE from three distinct tumor specimens. NT, no treatment; {alpha}2(IV)NC1, recombinant {alpha}2(IV)NC1 domain treated (2.5 mg/kg); control, control peptide treated (2.5 mg/kg). C: Co-localization (yellow) of apoptosis (green) and CD-31-positive (red) tumor blood vessels within control and {alpha}2(IV)NC1 domain-treated M21 tumors. Top: Representative examples of CD-31-positive blood vessels. Bottom: Representative examples of apoptosis in non-CD-31-expressing cells. Original magnifications: x200 (A); x630 (C).

 
Elevated Levels of Cellular Senescence in M21 Tumors from Mice Treated with Recombinant {alpha}2(IV)NC1 Domain

Integrin-mediated cellular interactions with the ECM are known to regulate a variety of processes involved in tumor growth, including cell adhesion, migration, apoptosis, proliferation, and cell cycle control. Interestingly, studies have suggested that chemotherapeutic agents and radiation therapy may modulate cellular senescence.38-41 In fact, cellular senescence has been observed in murine tumors and biopsies from human patients after chemotherapy.38-41 In this regard, we assessed whether {alpha}2(IV)NC1 domain could alter the levels of cellular senescence in tumors in vivo. To examine the relative levels of cellular senescence, we assessed the levels of SA-ß-Gal, a well-known marker for cellular senescence.27,28 Frozen sections of tumors from control and {alpha}2(IV)NC1 domain-treated mice were stained for SA-ß-Gal using X-Gal as previously described.27 As shown in Figure 7A , top, little if any cellular senescence was detected in tumors from control-treated mice. In contrast, tumors from mice treated with increasing doses of {alpha}2(IV)NC1 domain exhibited elevated levels of SA-ß-Gal-positive staining senescent cells scattered throughout the tumors (Figure 7A , middle and bottom). Importantly, tumors (n = 3) from mice treated with {alpha}2(IV)NC1 domain exhibited enhanced senescence staining (score +++) as compared to either tumors from untreated or control-treated mice (score +). Because inhibition of angiogenesis may result in reduction in growth factors, oxygen, and other nutrients to the tumor, it is possible that the induction of cellular senescence may be an indirect result of {alpha}2(IV)NC1 domain-inhibiting angiogenesis. Alternatively, {alpha}2(IV)NC1 domain may directly induce senescence in tumor cells.



View larger version (29K):
[in this window]
[in a new window]
  		Figure 7. Recombinant {alpha}2(IV)NC1 domain enhances tumor cell senescence. Tumors (M21) from control and {alpha}2(IV)NC1 domain-treated mice were snap-frozen and 4.0-µm tissue sections prepared. Tumor senescence was assessed by monitoring the expression of SA-ß-Gal. A: Representative examples of SA-ß-Gal expression (blue) within control and {alpha}2(IV)NC1 domain-treated M21 melanoma tumors. B: Representative examples of SA-ß-Gal expression (blue) within control and {alpha}2(IV)NC1 domain (250 µg/ml)-treated M21 cells cultured on collagen type-IV. C: Quantification of the relative numbers of senescence cells within control and {alpha}2(IV)NC1 domain-treated M21 melanoma cells cultured on collagen type-IV. Data bars represent the mean cell counts per x200 microscopic fields ± SE from triplicate wells. Experiments were completed three times with similar results. Original magnifications: x200 (A); x150 (B).

 
To determine whether {alpha}2(IV)NC1 domain can directly induce senescence in tumor cells, in vitro assays were performed with M21 melanoma cells cultured on a collagen type-IV substrate. M21 cells in the presence or absence of {alpha}2(IV)NC1 domain or control peptide were seeded in microtiter wells coated with collagen type-IV. The cells were allowed to incubate for 24 hours and then stained for the presence of SA-ß-Gal. As shown in Figure 7B (top), scattered SA-ß-Gal-positive staining senescent cells were detected in untreated cultures growing on collagen type-IV. Interestingly, M21 cells treated with exogenous {alpha}2(IV)NC1 domain exhibited an enhancement of SA-ß-Gal staining senescent cells after a 24-hour incubation period (Figure 7B , middle). In contrast, cells cultured under identical conditions treated with a control collagen peptide showed only minimal senescence staining (Figure 7B , bottom). In fact, {alpha}2(IV)NC1 domain caused a significant (P = 0.038) increase in tumor cell senescence by approximately threefold as compared to control (Figure 7C) . Importantly, recombinant {alpha}2(IV)NC1 domain failed to induce senescence in cells cultured on fibronectin (data not shown). These novel findings suggest for the first time, that recombinant {alpha}2(IV)NC1 domain may induce M21 tumor cell senescence and that this induction of tumor cell senescence is substrate-dependent.


   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
An expanding body of evidence has been accumulating concerning the requirement of cellular interactions with ECM components such as collagen in the regulation of invasive cellular processes.2-5,7 In fact, modulation of collagen structure and integrity has been suggested to alter integrin-mediated cell cycle control.42-45 Moreover, genetic evidence from collagen- and integrin-deficient mice has implicated these molecules as important regulators in biological events including proliferation, invasion, tumor growth, and angiogenesis.13,44 Previous studies have demonstrated that not only is cellular interactions with triple helical collagen important, but integrin-mediated ligation of cryptic epitopes hidden within its triple helical structure may alter integrin signaling and play crucial roles in angiogenesis and tumor growth.20,22 Thus, it is critical that we obtain a more complete understanding of the role that the ECM plays in regulating both normal tissue homeostasis and pathological diseases.

Recent work has provided convincing evidence that {alpha}2(IV)NC1 domain can inhibit endothelial cell migration and proliferation, and induce endothelial but not tumor cell apoptosis in vitro.14,35-37 Importantly, our studies confirm these findings because recombinant {alpha}2(IV)NC1 domain failed to induce tumor cell apoptosis in vitro yet did induce apoptosis in vivo. Studies have also suggested that specific NC domains can down-regulate FLIP, inhibit the phosphorylation of Akt and Fak, and disrupt mTOR function, which may play a role in their capacity to induce apoptosis in endothelial cells.14-19,35-37 However, little is known concerning the potential effect of {alpha}2(IV)NC1 domain on tumor cell behavior. Here, we demonstrate that recombinant {alpha}2(IV)NC1 domain can mediate tumor cell adhesion in an integrin-dependent manner using {alpha}vß3 and {alpha}1ß1.

To further expand our previous work, we examined the effects of recombinant {alpha}2(IV)NC1 domain on the growth of melanoma tumors in murine tumor models. Recombinant {alpha}2(IV)NC1 domain significantly inhibited human M21 melanoma tumor growth in a dose-dependent manner with a maximal inhibition (70%) observed at ~2.5 mg/kg. However, a common limitation associated with xenograft models is that they often do not mimic the tissue microenvironment in which the tumors are typically found in humans. In this regard, cutaneous human melanoma is thought to arise in part, from a series of genetic mutations within melanocytes present in the basal layers of the epidermis.46 Our findings suggest that daily (intraperitoneal) administration of recombinant {alpha}2(IV)NC1 domain potently inhibited human M21 melanoma tumor growth within full thickness human skin transplanted on nude mice. These findings confirm our previous results and suggest for the first time that {alpha}2(IV)NC1 domain has the capacity to inhibit human melanoma tumor growth within a human tissue microenvironment.

Although data from our laboratory and previously published studies from other investigators, suggest that recombinant {alpha}2(IV)NC1 domain can inhibit angiogenesis and tumor growth, the molecular mechanisms by which it functions is still not completely understood. Recent work from a variety of laboratories has provided compelling evidence that the inhibitory response in tumors to therapeutic intervention (chemotherapy and radiation) may be associated in part with the induction of cellular senescence.38-47,48 Senescence can be thought of as a physiological or cellular program of growth arrest. Cellular senescence can be triggered by a number of events including telomere shortening, DNA damage, and by other cellular stresses induced by chemotherapeutic reagents and radiation.38-40,46 Typical characteristics of senescent cells include an enlarged and flattened morphology with an increase in granularity. In addition, important biochemical changes that have been linked with cellular senescence include, increased expression of p21Waf1/CIP-1, and p16ink4a, an enhanced nuclear accumulation of actin and the expression of SA-ß-Gal.38-47,48 In fact, expression of SA-ß-Gal is used extensively as a reliable indicator of cellular senescence in vitro and in vivo.38-47,48 Importantly, although senescent cells do not grow, they remain metabolically active and have the potential to produce inhibitors and/or stimulators of angiogenesis and tumor growth.36-38,44 Thus, the net therapeutic benefit from the induction of cellular senescence may depend on the overall balance between expression of any inducers or inhibitors. Recent studies have suggested that certain tumors from murine models and patients treated with chemotherapy exhibit enhanced levels of cellular senescence.36,38,44 Consistent with these observations, our studies indicate a specific increase in the number of enlarged flattened tumor cells expressing SA-ß-Gal after treatment with {alpha}2(IV)NC1 domain in vitro. Importantly, this increase in relative levels of SA-ß-Gal-expressing senescent cells was not observed when cells were cultured on fibronectin, consistent with an integrin- and/or ECM-specific response. In this regard, it is interesting to note that changes in integrin-mediated tumor cell interactions with collagen have been reported to alter cell cycle progression.41,42 Moreover, alteration in Ras/MAP kinase-signaling pathways has also been shown to modulate cellular senescence.48 Thus, it would be interesting to speculate that the induction of tumor cell senescence observed after {alpha}2(IV)NC1 domain treatment may be associated in part with the alterations in integrin-mediated tumor cell interactions with collagen. Given this possibility, it would also be interesting to speculate that specific tumors with distinct integrin profiles such as those that express {alpha}vß3 and {alpha}1ß1 might be more responsive to recombinant {alpha}2(IV)NC1 domain than those tumors lacking these integrins. Studies are currently underway to examine this intriguing possibility.

Taken together, our findings suggest that although {alpha}2(IV)NC1 domain likely impacts tumor growth via an anti-angiogenic mechanisms involving induction of endothelial cell apoptosis, it may also directly affect tumor cell adhesion and senescence via an integrin- and/or ECM-specific mechanism. The dual effects of recombinant {alpha}2(IV)NC1 domain on both angiogenesis and tumor cells have important clinical implications. For example, recent studies have provided evidence that anti-angiogenic drugs such as vascular endothelial growth factor antagonists, in conjunction with conventional chemotherapy can enhance clinical tumor response in human trials.49 Moreover, other studies suggest that induction of cellular senescence in tumors may enhance tumor sensitivity to conventional chemotherapy as well as reverse drug resistance in certain tumors.38-40 Our findings that {alpha}2(IV)NC1 domain can inhibit tumor cell adhesion and induce tumor cell senescence, in conjunction with its previously reported anti-angiogenic activity, provide support for the notion that the use of {alpha}2(IV)NC1 domain alone or in combination with standard chemotherapy and/or radiotherapy may represent an important new approach for the treatment of malignant human tumors.


   Acknowledgements
 
We thank Sharon Binns for her help in preparing the manuscript.


   Footnotes
 
Address reprint requests to Peter C. Brooks, Ph.D., New York University School of Medicine, Departments of Radiation Oncology and Cell Biology, The NYU Cancer Institute, Rusk Bldg. Rm. 806, 400 East 34th St., New York, NY 10016. E-mail: peter.brooks{at}med.nyu.edu

Supported in part by BioStratum Inc. and the National Institutes of Health (grant RO1-CA91645 to P.C.B).

Dr. Brooks is a paid consultant for BioStratum, Inc.

Accepted for publication November 30, 2004.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Gagne P, Akalu A, Brooks PC: Challenges facing anti-angiogenic therapy for cancer: impact of the tumor extracellular environment. Expert Rev Anticancer Ther 2004, 4:129-140[Medline]
  2. Stupack DG, Cheresh DA: Apoptotic cues from the extracellular matrix: regulators of angiogenesis. Oncogene 2003, 22:9022-9029[Medline]
  3. Schenk S, Quaranta V: Tales from the cryptic sites of the extracellular matrix. Trends Cell Biol 2003, 13:366-375[Medline]
  4. Kalluri R: Basement membranes: structure, assembly and role in tumor angiogenesis. Nat Rev Cancer 2003, 3:422-433[Medline]
  5. Ahmad SA, Jun YD, Liu W, Reinmuth N, Parikh A, Stoeltzin O, Fan F, Elli LM: The role of the microenvironment and intercellular cross-talk in tumor angiogenesis. Semin Cancer Biol 2002, 12:105-112[Medline]
  6. Brooke BS, Karnik SK, Li DY: Extracellular matrix in vascular morphogenesis and disease: structure versus signal. Trends Cell Biol 2003, 13:51-56[Medline]
  7. Ortega N, Werb Z: New functional roles for non-collagenous domains of basement membrane collagens. J Cell Sci 2002, 115:42014214
  8. Myllyharju J, Kivirkko KI: Collagens, modifying enzymes and their mutations in human, flies and worms. Trends Cell Biol 2004, 20:33-43
  9. Zhang X, Hudson BG, Sarras MP: Hydra cell aggregate development is blocked by selective fragments of fibronectin and type IV collagen. Dev Biol 1994, 164:10-23[Medline]
  10. Sarras M, Hudson B: Use of Isolated Domains of Type-IV Collagen to Modify Cell and Tissue Interactions. 1997 Government Printing Office, Washington D.C. issued patent no. 5,691,182
  11. Sundaramoorthy M, Meiyappan M, Todd P, Hudson BG: Crystal structure of NC1 domains: structural basis for type-IV collagen assembly in basement membranes. J Biol Chem 2002, 277:31142-31153[Abstract/Free Full Text]
  12. Emsley J, Knight CG, Farndale RW, Barnes MJ, Liddington RC: Structural basis of collagen recognition by integrin {alpha}2ß1. Cell 2000, 100:47-56
  13. Hynes RO: Integrins: bidirectional, allosteric signaling machines. Cell 2002, 110:673-687[Medline]
  14. Petitclerc E, Boutaud A, Prestakyo A, Xu J, Sado Y, Ninomiya Y, Sarras MP, Hudson BG, Brooks PC: New functions for non-collagenous domains of human collagen type-IV: novel integrin ligands inhibiting angiogenesis and tumor growth in vivo. J Biol Chem 2000, 275:8051-8061[Abstract/Free Full Text]
  15. Sudhakar A, Sungimoto H, Yang C, Lively J, Zeisberg M, Kalluri R: Human tumstatin and human endostatin exhibit distinct antiangiogenic activities mediated by {alpha}vß3 and {alpha}5ß1 integrins. Proc Natl Acad Sci USA 2003, 100:4755-4771
  16. Floquet N, Pasco S, Ramont L, Derreumaux P, Larzoes JY, Nuzillard JM, Maquart FX, Alix AJP, Monboisse JC: The antitumor properties of the {alpha}3(IV)-(185-203) peptide from the NC1 domain of type-IV collagen (tumstatin) are conformation-dependent. J Biol Chem 2004, 279:2091-2100[Abstract/Free Full Text]
  17. Colorado PC, Torre A, Kamphaus G, Maeshima Y, Hopfer H, Takahashi K, Volk R, Zamborshy ED, Merman S, Sarkar PK, Ericksen MB, Dhanabal M, Simon M, Post M, Kufe DW, Weichselbaum RR, Sukhatme VP, Kalluri R: Anti-angiogenic cues from the vascular basement membrane collagen. Cancer Res 2000, 60:252-256
  18. Pedchenko V, Zent R, Hudson BG: {alpha}vß3 and {alpha}vß5 integrins bind both the proximal RGD site and non-RGD motifs within noncollagenous (NC1) domain of the {alpha}3 chain of type-IV collagen. J Biol Chem 2004, 279:2772-2780[Abstract/Free Full Text]
  19. Maeshima Y, Colorado PC, Torre A, Holthaus KA, Grunkemeyer JA, Ericksen MB, Hopfer H, Xiao Y, Stillman IE, Kalluri R: Distinct antitumor properties of a type IV collagen domain derived from basement membrane. J Biol Chem 2000, 275:2134-2138
  20. Hangai M, Kitaya N, Xu J, Chan CK, Kim JJ, Werb Z, Ryan SJ, Brooks PC: Matrix metalloproteinase-9-dependent exposure of a cryptic migratory control site in collagen is required before retinal angiogenesis. Am J Pathol 2002, 161:1429-1437[Abstract/Free Full Text]
  21. Brooks PC, Silletti S, von Schalscha TL, Friedlander M, Cheresh DA: Disruption of angiogenesis by PEX, a noncatalytic metalloproteinase fragment with integrin binding activity. Cell 1998, 92:391-400[Medline]
  22. Xu J, Rodriguez D, Petitclerc E, Kim JJ, Hangai M, Moon SM, Davis GE, Brooks PC: Proteolytic exposure of a cryptic site within collagen type-IV is required for angiogenesis and tumor growth in vivo. J Cell Biol 2001, 154:1069-1079[Abstract/Free Full Text]
  23. Petitclerc E, Stromblad S, von Schalscha TL, Mitjans F, Piulats J, Montgomery AM, Cheresh DA, Brooks PC: Integrin {alpha}vß3 promotes M21 melanoma growth in human skin by regulating tumor cell survival. Cancer Res 1999, 59:2724-2730[Abstract/Free Full Text]
  24. Gasparini G, Brooks PC, Biganzoli E, Vermeulen PB, Bonoldi E, Dirix LY, Ranieri G, Miceli R, Cheresh DA: Vascular integrin {alpha}vß3: a new prognostic indicator in breast cancer. Clin Cancer Res 1998, 11:2625-2534
  25. Matkowsky K, Glover S, Benya R: Quantitative immunohistochemistry: a new algorithm measuring cumulative signal strength and receptor number. Micro Anal 2004, 18:5-6
  26. Brooks PC, Montgomery AM, Rosenfeld M, Reisfeld RA, Hu T, Klier G, Cheresh DA: Integrin {alpha}vß3 antagonists promote tumor regression by inducing apoptosis of angiogenic blood vessels. Cell 1994, 79:1157-1164[Medline]
  27. Dimri GP, Lee X, Basile G, Acosta M, Scott G, Roskelly C, Medranno EE, Linskens M, Rubelj I, Pereira-Smith O, Peacocke M, Campisi J: A biomarker that identifies senescent human cells in culture and in aging skin in vivo. Proc Natl Acad Sci USA 1995, 92:9363-9367[Abstract/Free Full Text]
  28. Zheng X, Chou PM, Mirkin BL, Rebbaa A: Senescence-initiated reversal of drug resistance: specific role of cathepsin L. Cancer Res 2004, 64:1773-1780[Abstract/Free Full Text]
  29. Han J, Ohno N, Passco S, Monboisse JC, Borel JP, Kefalides NA: A cell binding domain from the {alpha}3 chain of type-IV collagen inhibits proliferation of melanoma cell. J Biol Chem 1997, 272:20395-20401[Abstract/Free Full Text]
  30. Lee JW, Juliano R: Mitogenic signal transduction by integrin- and growth factor receptor-mediated pathways. Mol Cells 2004, 17:188-202[Medline]
  31. Guo W, Giancottti FG: Integrin signaling during tumor progression. Nat Rev Mol Cell Biol 2004, 10:816-826
  32. Brooks PC: Role of integrins in angiogenesis. Eur J Cancer 1996, 32A:2423-2429
  33. Brooks PC, Stromblad S, Klemke R, Visscher D, Sarkar FH, Cheresh DA: Anti-integrin {alpha}vß3 blocks human breast cancer growth and angiogenesis in human skin. J Clin Invest 1995, 96:1815-1822
  34. Brooks PC, Clark RA, Cheresh DA: Requirement of vascular integrin {alpha}vß3 for angiogenesis. Science 1994, 264:569-571[Abstract/Free Full Text]
  35. Folkman J: Role of angiogenesis in tumor growth and metastasis. Semin Oncol 2002, 29:15-18[Medline]
  36. Kamphaus GD, Colorado PC, Panka DJ, Hopfer H, Ramchandran R, Torres A, Maeshima Y, Mier JW, Sukhatme VP, Kalluri R: Canstatin, a novel matrix-derived inhibitor of angiogenesis and tumor growth. J Biol Chem 2000, 275:1209-1215[Abstract/Free Full Text]
  37. Panka DJ, Mier JW: Canstatin inhibits Akt activation and induces Fas-dependent apoptosis in endothelial cells. J Biol Chem 2003, 278:37632-37636[Abstract/Free Full Text]
  38. He GA, Luo JX, Zhang TY, Wang FY, Li RF: Canstatin-N fragment inhibits in vitro endothelial cell proliferation and suppresses in vivo tumor growth. Biochem Biophys Res Commun 2003, 312:801-805[Medline]
  39. Roninson IB: Tumor cell senescence in cancer treatment. Cancer Res 2003, 63:2705-2715[Abstract/Free Full Text]
  40. Schmitt CA, Fridman JS, Yang M, Lee S, Baranov E, Hoffman RM, Lowe SW: A senescence program controlled by p53 and p16INK4a contributes to the outcome of cancer therapy. Cell 2002, 109:335-346[Medline]
  41. Poele RH, Okorokov AL, Jardine L, Cummings J, Joel SP: DNA damage is able to induce senescence in tumor cells in vitro and in vivo. Cancer Res 2002, 62:1876-1883[Abstract/Free Full Text]
  42. Henriet P, Zhong ZD, Brooks PC, Weinberg KI, DeClerck YA: Contact with fibrilla collagen inhibits melanoma cell proliferation by up-regulating p27KIP1. Proc Natl Acad Sci USA 2000, 97:10026-10031[Abstract/Free Full Text]
  43. Hotary KB, Allen ED, Brooks PC, Datta NS, Long MW, Weiss SJ: Membrane type-1 matrix metalloproteinase usurps tumor growth control imposed by three-dimensional extracellular matrix. Cell 2003, 114:33-45[Medline]
  44. Fringer J, Grinnell F: Fibroblast quiescence in floating collagen matrices. J Biol Chem 2003, 278:20612-20617[Abstract/Free Full Text]
  45. Gustafsson E, Fassler R: Insight into extracellular matrix functions from mutant mouse models. Exp Cell Res 2000, 261:52-68[Medline]
  46. Perlis C, Herlyn M: Recent advances in melanoma biology. Oncologist 2004, 9:182-187[Abstract/Free Full Text]
  47. Chang BD, Swift ME, Shen M, Fang J, Broude EV, Roninson IB: Molecular determinants of terminal growth arrest induced in tumor cells by a chemotherapeutic agent. Proc Natl Acad Sci USA 2002, 99:389-394[Abstract/Free Full Text]
  48. Lin AW, Barradas M, Stone JC, Aelst L, Serrano M, Lowe SW: Premature senescence involving p53 and p16 is activated in response to constitutive MEK/MAPK mitogenic signaling. Genes Dev 1998, 12:3008-3019[Abstract/Free Full Text]
  49. Yang JC, Haworth L, Sherry RM, Hwu P, Schwartzentruber DJ, Topalian SL, Steinberg SM, Chen HX, Rosenberg SA: A randomized trial of bevacizumab, an anti-vascular endothelial growth factor antibody for metastatic renal cancer. N Engl J Med 2003, 349:427-434[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
D. Barkan, H. Kleinman, J. L. Simmons, H. Asmussen, A. K. Kamaraju, M. J. Hoenorhoff, Z.-y. Liu, S. V. Costes, E. H. Cho, S. Lockett, et al.
Inhibition of Metastatic Outgrowth from Single Dormant Tumor Cells by Targeting the Cytoskeleton
Cancer Res., August 1, 2008; 68(15): 6241 - 6250.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
L. G. Wang, E. M. Johnson, Y. Kinoshita, J. S. Babb, M. T. Buckley, L. F. Liebes, J. Melamed, X.-M. Liu, R. Kurek, L. Ossowski, et al.
Androgen Receptor Overexpression in Prostate Cancer Linked to Pur{alpha} Loss from a Novel Repressor Complex
Cancer Res., April 15, 2008; 68(8): 2678 - 2688.
[Abstract] [Full Text] [PDF]

Home page
 J. Am. Soc. Nephrol.Home page
C. M. Borza, D.-B. Borza, V. Pedchenko, M. A. Saleem, P. W. Mathieson, Y. Sado, H. M. Hudson, A. Pozzi, J. Saus, D. R. Abrahamson, et al.
Human Podocytes Adhere to the KRGDS Motif of the {alpha}3{alpha}4{alpha}5 Collagen IV Network
J. Am. Soc. Nephrol., April 1, 2008; 19(4): 677 - 684.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
C. S. Boosani, A. P. Mannam, D. Cosgrove, R. Silva, K. M. Hodivala-Dilke, V. G. Keshamouni, and A. Sudhakar
Regulation of COX-2 mediated signaling by {alpha}3 type IV noncollagenous domain in tumor angiogenesis
Blood, August 15, 2007; 110(4): 1168 - 1177.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
A. Cretu, J. M. Roth, M. Caunt, A. Akalu, D. Policarpio, S. Formenti, P. Gagne, L. Liebes, and P. C. Brooks
Disruption of Endothelial Cell Interactions with the Novel HU177 Cryptic Collagen Epitope Inhibits Angiogenesis
Clin. Cancer Res., May 15, 2007; 13(10): 3068 - 3078.
[Abstract] [Full Text] [PDF]