A novel technique for quantifying changes in vascular density, endothelial cell proliferation and protein expression in response to modulators of angiogenesis using the chick chorioallantoic membrane (CAM) assay
- Walter J Miller†1,
- Mark L Kayton1,
- Angela Patton2,
- Sarah O'Connor1,
- Mei He1,
- Huan Vu1,
- Galina Baibakov3,
- Dominique Lorang1,
- Vladimir Knezevic3,
- Elise Kohn2,
- H Richard Alexander1,
- David Stirling4,
- Faribourz Payvandi4,
- George W Muller4 and
- Steven K Libutti†1Email author
© Miller et al; licensee BioMed Central Ltd. 2004
Received: 22 November 2003
Accepted: 30 January 2004
Published: 30 January 2004
Reliable quantitative evaluation of molecular pathways is critical for both drug discovery and treatment monitoring. We have modified the CAM assay to quantitatively measure vascular density, endothelial proliferation, and changes in protein expression in response to anti-angiogenic and pro-angiogenic agents. This improved CAM assay can correlate changes in vascular density with changes seen on a molecular level. We expect that these described modifications will result in a single in vivo assay system, which will improve the ability to investigate molecular mechanisms underlying the angiogenic response.
The application of antiangiogenic agents for cancer therapy may improve the results of conventional anticancer treatments because of increased tumor specificity and decreased development of tumor cell resistance [1, 2]. The correlation between increased angiogenesis and poor outcome has been established in breast cancer, renal cancer, prostate cancer, colon cancer and melanoma . Recent clinical trials have provided support for the use of antiangiogenic agents in oncology. Yang et al. have shown that single agent bevacizumab (anti-VEGF antibody, Avastin) can significantly prolong the time to progression of disease in patients with metastatic renal-cell cancer . In addition, Giantonio et al. have demonstrated the benefit of the combination of bevacizumab and chemotherapy as treatment for advanced colorectal cancer resulting in improved survival . The balance between angiogenesis and antiangiogenesis involves several mechanisms and pathways in addition to the modulation of VEGF. Identification and validation of these additional factors will provide new targets for antiangiogenic strategies . In order to capitalize on these discoveries, improved assay systems are needed to screen for novel angiogenesis inhibitors and to elucidate their mechanism of action.
Angiogenesis is a complex process that involves an appropriate environment of growth factors, extracellular matrix proteins, proteases, and cell surface integrins to be present so that the cellular events of adhesion, proteolysis, migration, invasion and proliferation can occur . This interplay of host factors is critical for angiogenesis to proceed. Any quantitative assay system needs to recapitulate these components as closely as possible. A variety of in vitro and in vivo angiogenesis assay systems have been developed to measure the effect of various agents on endothelial cell proliferation and blood vessel growth . Many of these are subjective and qualitative with limited quantitative potential, and result in variable data from one laboratory to another. Most are based on isolated cell preparations, which do not adequately represent the complex host components and interactions that occur in vivo. Though in vivo small animal models are a closer approximation to the processes seen in humans, they can be time consuming and expensive especially during the early stages of agent evaluation and screening. An alternative to small animal models is the chick chorioallantoic membrane (CAM) which provides a natural environment of growing blood vessels and all the components of the complex host interaction . However, its utility has been limited due to difficulty in measuring the angiogenic response to an experimental compound in an objective and quantifiable manner, and by the inability to identify the molecular basis for these changes.
By utilizing novel methods to quantify vascular density and protein expression, we have adapted the CAM assay to create an in vivo angiogenesis model system that is rigorously quantitative, amenable to high-throughput screening, and applicable for the testing of systemic and/or topical administration of experimental agents. We have validated this assay with two known inhibitors of angiogenesis, known to modulate molecular proteins, and have then used the assay to identify and partially characterize a novel angiogenesis inhibitor.
Materials and Methods
Chick embryos were acquired on embryonic day 10 from CBT Farms (Chestertown, Maryland) from White Leghorn chickens and incubated (RX2 Incubator, Lyon Electric, Chua Vista, California) at 98.6 degrees F and 51 % relative humidity.
Preparation of the filter disks
Small filter disks (Whatman filter paper #1001 090) were generated using a standard 6 mm hole puncher and sterilized by autoclaving. Two milliliters of Cortisone acetate solution (3 mg/ml in 95%ethanol; Sigma#C3130) was evenly pipetted as a thin layer over the filter disks until saturated and allowed to air-dry under a laminar flow hood.
The CAM assay was performed in an identical fashion as described by Brooks et al. . Briefly, day 10 eggs were candled using a hand held egg candler at the blunt end of the egg to identify the air sac and prominent blood vessels. Using a Dremel model drill (Dremel #750 MiniMite), the CAM was separated from the shell by making a shallow burr hole at the blunt end on the egg and another burr hole made perpendicular to the previously identified blood vessels in the center of the egg. Mild suction was applied to the blunt end burr hole to displace the air sac and drop the CAM away from the shell. Fine forceps were then used to pick away the shell over the false air sac, so that a window could be made and the CAM identified. Next, 15 μl of recombinant human b-FGF(R&D Systems, #233-FB/CF) 2 ug/ml or vehicle (0.1%BSA/PBS) was pipetted onto a cortisone dried filter disk and the disk then placed on the CAM in an avascular area. The window was sealed with sterile Scotch tape and the egg returned to the incubator.
Systemic Injection of Antagonists
At day 11, a small hole over the site of the previously identified blood vessels was made using a razor blade. The overlying shell was removed leaving the underlying white shell membrane intact. A drop of mineral oil was used to render the underlying blood vessels transparent. Using a 30-gauge needle, 100 μl of 100 μM fumagillin in 0.1% DMSO (Sigma-Aldrich, #F6771), 100 μl of 3 mg/ml LM609 (Chemicon Inc., Temecula, California), 100 μl of 0.1 μM CC5079 (Celgene Corp., Warren, New Jersey) or 100 μl of a carrier solution (0.1% DMSO) was injected into the blood vessel. Blood seepage was controlled with a cotton tip applicator, the window covered with sterile Scotch tape and the egg placed back into the incubator for 48 hours.
CAM fluorescent confocal microscopy
At day 13, a second window was made on the opposite side of the egg from the original systemic injection site. A moderately sized blood vessel was identified and injected with FITC-dextrin 1 mg/ml (Cat#FD-2000S, Sigma) and placed back into the incubator for 30 minutes. The CAM disks were then harvested and placed into 10% formalin and stored at +4 degrees C. The CAM disk specimens were harvested for light photomicroscopy and then viewed under the Zeiss LSM 5 Pascal confocal fluorescent microscope using the FITC filter at 488 nm. The image was then analyzed by the Pascal imaging software (Version 2.8 SP1) so that the mean fluorescent vascular density could be measured. Images were then filtered to remove low level background and large parent vessel interference so that only new vessel blood formation pixel intensity could be measured.
Standard Blinded Grading
The CAM disks were harvested for light photomicroscopy and given to three blinded graders for scoring. Scoring was based from 0 – 4 (0 = no angiogenesis present in CAM, 4 = strong angiogenesis reaction in CAM). Results of three blinded graders for each group were then scored.
CAM-XTT assay analysis
Twenty-four hours after systemic injection of an inhibitor or carrier, day 12 eggs were removed from the incubator so that they could undergo a cell proliferation assay via a XTT cell proliferation assay kit (Roche Applied Science Cat#1465015). Five ml of XTT labeling reagent and 0.1 ml electron coupling reagent were mixed together and 30 μl applied via pipette onto the CAM disk. A small piece of sterile Scotch tape was reapplied over the central window and the eggs were returned to the incubator for 24 hours. At day 13 the disks were harvested for light photomicroscopy and the mean spectrophotometric absorbance at 450 nm was measured (reference wavelength 620 nm (Titertek Multiskan Ascent Plate Reader)).
CAM-Layered Expression Scanning (LES)
After systemic injection of the experimental compound or control vehicle, day 13 fertilized eggs were removed from the incubator. The CAM specimens were harvested, placed into 70% ethanol and stored at -80 degrees C. Layered Expression Scanning (LES) was performed on the CAM specimens according to 20/20 Gene Systems' protocols (20/20 Gene Systems, Gaithersburg, Maryland). After transfer, total protein bound to the membranes was biotinylated, followed by immuno-detection of the specific protein of interest. At the end of the assay, biotinylated total protein was visualized by streptavidin/FITC conjugate and specific protein/primary antibody complexes were visualized by appropriate secondary antibody conjugated to Cy3. Fluorescence on the membranes was detected on a Scan Array Express Microarray Scanner (Packard Biosciences) and measured by using ScanArray Express Software (Packard Biosciences). LES used the following antibodies: rabbit anti-MetAP-2 (Zymed Laboratories Inc., 1:100 dilution, Cat#71–7200), rabbit anti-p21 (Zymed Laboratories Inc., 1:100 dilution, Cat#71–1000), mouse anti-human integrin alpha v beta 3 (Chemicon Intl., 1:500 dilution, Cat# MAB1976B), and rabbit anti-caspase-3 (Cell Signaling Technology, 1:500 dilution, Cat# 9661). Protein/ primary antibody complexes were detected with appropriate secondary antibodies conjugated to Cy3 and analyzed on a Scan Array Express Microarray Scanner (Packard Biosciences).
All statistical analyses were performed using InStat software (GraphPad Software Inc., San Diego, California) on a PowerBook G4 (Apple Computer Inc, Cupertino, California). Student's t test was performed when the standard deviations between groups were found to be equal. An Alternate Welch t test or nonparametric Mann-Whitney U test was performed when the standard deviations between groups were not equal.
Fluorescent confocal microscopy imaging in the chorioallantoic membrane assay
XTT cell proliferation assay in the CAM assay
Layered Expression Scanning (LES) in the CAM assay to quantify changes in protein expression and vascular density at the molecular level
We specifically probed the membranes for proteins known to be involved in the mechanism of action of fumagillin and LM609: p21, MetAP-2 and alpha v beta3. In addition, we probed the blots for proteins involved in the apoptotic cascade. Our results demonstrated that the amount of p21 was significantly increased (P < 0.0001, Student's t test) in CAM disks treated with fumagillin (Figure 4C). MetAP-2 was found to be significantly increased in CAM disks treated with fumagillin (P = 0.0042, Alternate Welch t test) and LM609 (P = 0.0253, Alternate Welch t test), Figure 4D). This data is consistent with the published literature, which has demonstrated that both p21 and MetAP-2 are increased in endothelial cells in response to fumagillin exposure [12, 13].
Evaluation of a novel putative antiangiogenic compound
Having validated this new method for quantifying endothelial cell proliferation in the CAM assay using two known angiogenesis inhibitors, we next utilized the XTT assay to evaluate a putative small molecule inhibitor of angiogenesis, CC5079. Work in our laboratory has demonstrated that CC5079 inhibits endothelial cell proliferation in vitro in a dose dependent manner (data not shown). In order to determine if CC5079 had antiangiogenic properties in vivo, we treated day 10 eggs containing bFGF stimulated disks with either systemic CC5079 (0.1 μM) or a control carrier vehicle (0.1% DMSO) 24 hours after disk placement. The XTT reagent was then applied topically to the disks 24 hours after systemic injection and the disks were scored for endothelial cell proliferation using a plate reader. Significant inhibition of endothelial proliferation was detected in vivo as a result of systemic delivery of CC5079 using this assay (Figure 3B).
As antiangiogenic agents begin to show activity in the clinic, there is increasing interest in angiogenesis inhibition as a viable means for treating cancer [4, 5]. In order to improve upon these successes, as well as to help guide the choice of therapeutic combinations, a better understanding of the mechanisms of action and relative degree of activity of these agents is important. Robust screening methods are needed which can help in the selection of compounds that look promising for further evaluation. Our modifications of the in vivo CAM assay may provide a useful tool for this purpose.
The creation of a fluorescent angiogram of the CAM is a simple modification which renders the assay more quantitative in the analysis of vascular density than previously described . In this study, differences between treatment groups could be detected when CAM disks were scanned under fluorescent confocal microscopy (Figure 2B) and the images analyzed with image analysis software (LSM 5 Pascal Software, Version 2.8 SP1, Copyright Carl Zeiss 1986–2000). CAM disks stimulated with bFGF (Figure 1E and 1H) had a greater degree of angiogenesis than unstimulated disks (Figure 1D, and 1G) or disks that were stimulated but then treated with systemic injection of fumagillin (Figure 1F and 1I). These differences were missed by blinded graders (Figure 2A) using standard scoring systems due to the variability in vessel detection using this method. When comparing the light photomicrographs and FITC angiograms of stimulated CAM disks, it is apparent that many small blood vessels are not detected by standard microscopy (Figure 1A,1B). The use of fluorescent confocal imaging and FITC-dextran injection can render an obscure CAM disk readable and allows for the quantitative measurement of fluorescent vascular density. Of note, we have also used FITC labeled lectin in similar experiments with equivalent results (data not shown).
In order to directly quantify changes in endothelial cell proliferation in vivo, we applied the XTT reagent to the CAM assay. The XTT assay is based on the cleavage of the yellow tetrazolium salt XTT to form an orange formazen dye by metabolically active cells. In our experiments, CAM disks treated with bFGF and the systemically administered angiogenesis inhibitors, fumagillin or LM609, showed significant inhibition of endothelial cell proliferation by the XTT assay as measured by a decrease in overall mean absorbance (Figure 3A).
We also tested the putative angiogenesis inhibitor, CC5079, for activity using the XTT modified CAM assay. CC5079 is an orally bioavailable agent with in vitro effects on endothelial cells including inhibition of cytokine release, inhibition of proliferation, and inhibition of tubulin polymerization. CC5079 showed a significant decrease in mean absorbance in the XTT assay when compared to untreated bFGF stimulated CAM disks (Figure 3B). Use of the XTT assay may be a useful method to quantify endothelial cell proliferation or inhibition in the CAM assay in a more direct fashion. Such a measurement is complementary to the measurements obtained using the FITC-dextran angiograms which quantify vascular density and can be used to study vessel architecture and branching.
In developing this new assay system we feel it was important not only to be able to quantify the phenotypic changes in angiogenesis, but also to detect proteonomic changes responsible for these effects. Layered Expression Scanning (LES) has been previously used to perform protein analysis and quantification in tumor samples [17, 18]. This technique is based on removing proteins from the tissue samples and presenting them to a thin absorbent membrane for detection and identification. The technique is similar to a Western Blot in that the proteins are more bioavailable to subsequent antibody analysis than they might otherwise be using approaches such as immunohistochemistry. In addition, LES allows for preservation of tissue architecture and for multiple blots to be created from a single sample.
To validate this technique, we investigated a pathway previously determined to play a role in the mechanism of action of fumagillin. Fumagillin has been shown to modulate p21 and methionine peptidase (MetAP-2) [12, 13]. Changes in the levels of expression of p21 and MetAP-2 have been shown to correlate with the effects of fumagillin on endothelial cell proliferation [19, 20]. Our results demonstrated that p21 staining was greatest in CAM disks stimulated with bFGF and treated with systemic fumagillin when compared to controls (Figure 4C). Furthermore, MetAP-2 staining was greatest in CAM disks stimulated with bFGF and treated with systemic fumagillin. These findings are consistent with the published literature and help to validate this approach [12, 13]. We also found that there was an up regulation of MetAP-2 in disks treated with LM609 (Figure 4D). The accumulation of the G-1 cyclin-dependent kinase inhibitor p21 is known to be caused by the activation of the p53 pathway after exposure to TNP-470, a fumagillin analog . The exact mechanism for the increase in MetAP-2 by fumagillin is unknown, however previous studies have shown that MetAP-2 acts as a direct molecular target for TNP-470 . The up regulation of MetAP-2 in association with LM609 has not been previously described.
There was a significant decrease in the amount of staining of alpha v beta 3 in CAM disks treated with fumagillin (Figure 5A). The mechanism of alpha v beta3 down regulation in CAM disks after systemic treated with fumagillin has not been previously described. Staining for cleaved caspase-3 was greatest in stimulated CAM disks treated with systemic carrier. A significantly lower amount of staining was seen in CAM disks treated with fumagillin and LM609 (Figure 5B). Both, fumagillin and LM609 are known to induce apoptosis. Methionine aminopeptidase (MetAP-2) is known to be the molecular target of fumagillin. Fumagillin has been shown to inhibit cancer cell proliferation and MetAP-2 may play a complex role in tumor cell progression. Previous studies have examined the expression and function of MetAP-2 in an in vitro model of human mesothelioma. Fumagillin was found to induce apoptosis due to early mitochondrial damage in malignant cells. Transfection of mesothelioma cells with MetAP-2 anti-sense oligonucleotide induced MetAP-2 inhibition and increased caspase activity and the caspase inhibitor, zVAD-fmk, preventing fumagillin-induced apoptosis . Therefore, MetAP-2 and caspase activity may be inversely related.
We also analyzed the effects of the putative angiogenesis inhibitor CC5079 after systemic injection in the CAM assay on the above pathways. Our results demonstrated that p21 was significantly increased by treatment with CC5079 compared to control. There was no significant effect on MetAP-2 expression compared to controls, while CC5079 increased expression of cleaved caspase-3 and inhibited the expression of alpha v beta 3 (Figure 6A,6B,6C,6D). These results suggest a possible role of CC5079 as an inducer of endothelial cell apoptosis perhaps through a MetAP-2 independent pathway. Additional studies will now be pursued based on these leads to further characterize the activity and mechanism of action of CC5079.
In conclusion, we have described a modified version of the CAM assay, which utilizes systemic injection of FITC-dextran to create a quantitative measurement of vascular density. In addition, XTT reagent can be utilized as a complementary method to quantitatively measure endothelial proliferation in the CAM after systemic or topical administration of an anti-angiogenic or pro-angiogenic agent. Finally, Layered Expression Scanning (LES) may be very useful as a method to quantitatively measure protein changes in response to anti-angiogenic and pro-angiogenic agents in the CAM assay and therefore help to correlate changes in vascular density with changes seen on a molecular level. It is our hope that these described modifications will result in a single in vivo assay system that will help investigators to better understand the molecular mechanisms underlying the angiogenic response following exposure to agents and may help in the identification of novel targets and the design of novel therapeutics.
- Millauer B, Shawver LK, Plate KH, Risau W, Ullrich A: Glioblastoma growth inhibited in-vivo by dominant-negative Flk-1mutant. Nature. 1994, 367: 576-579. 10.1038/367576a0.View ArticlePubMedGoogle Scholar
- Kim KJ, Li B, Winer J, Armanini M, Gillett N, Phillips HS, Ferrara N: Inhibition of vascular endothelial growth factor-induced angiogenesis suppresses tumor growth in-vivo. Nature. 1993, 362: 841-844. 10.1038/362841a0.View ArticlePubMedGoogle Scholar
- Gasparini G, Harris AL: Clinical importance of the determination of tumor angiogenesis in breast carcinoma: much more than a new prognostic tool. J Clin Oncol. 1995, 13: 765-782.PubMedGoogle Scholar
- Yang JC, Haworth L, Sherry RM, Hwu P, Schwartzentruber DJ, Topalian SL, Steinberg SM, Chen HX, Rosenberg SA: A Randomized trial of bevacizumab, an antivascular endothelial growth factor antibody for metastatic renal cancer. N Engl J Med. 2003, 349: 427-434. 10.1056/NEJMoa021491.PubMed CentralView ArticlePubMedGoogle Scholar
- Giantonio BJ, Levy D, O'Dwyer PJ, Meropol NJ, Catalano PJ, Benson B: Bevacizumab (anti-VEGF) plus IFL irinotecan, fluorouracil. leucovorin) as front line therapy for advanced colorectal cancer (advCRC): Results from the Eastern Cooperative Oncology Group (ECOG) Study E2200. ASCO. 2003, 1024:Google Scholar
- Toi M: Endothelial growth factors: Target of antiangiogenesis, Ovarian Cancer Research notebook. The Cancer Journal. 2003, 8: 1-7.Google Scholar
- Auerbach W, Auerbach R: Angiogenesis inhibition: a review. Pharmacol Ther. 1994, 63: 265-311. 10.1016/0163-7258(94)90027-2.View ArticlePubMedGoogle Scholar
- Auerbach R, Lewis R, Shinners B, Kubai L, Akhtar N: Angiogenesis assays: a critical overview. Clin Chem. 2003, 49: 32-40. 10.1373/49.1.32.View ArticlePubMedGoogle Scholar
- Ribatti D, Vacca A, Roncali L, Dammacco F: The chick embryo chorioallantoic membrane as a model for in vivo research on anti-angiogenesis. Curr Pharm Biotechnol. 2000, 1: 73-82.View ArticlePubMedGoogle Scholar
- Ingber D, Fujita T, Kishimoto S, Sudo K, Kanamaru T, Brem H, Folkman J: Synthetic analogues of fumagillin that inhibit angiogenesis and suppress tumor growth. Nature. 1990, 348: 555-557. 10.1038/348555a0.View ArticlePubMedGoogle Scholar
- Yasukawa T, Kimura H, Tabata Y, Miyamoto H, Honda Y, Ikada Y, Ogura Y: Targeted delivery of anti-angiogenic agent TNP-470 using water-soluble polymer in the treatment of choroidal neovascularization. Invest Ophthalmol Vis Sci. 1999, 40: 2690-2696.PubMedGoogle Scholar
- Zhang Y, Griffith EC, Sage J, Jacks T, Liu JO: Cell cycle inhibition by the anti-angiogenic agent TNP-470 is mediated by p53 and p21WAF1/CIP1. Proc Natl Acad Sci. 2000, 6: 6427-6432. 10.1073/pnas.97.12.6427.View ArticleGoogle Scholar
- Wang J, Lou P, Henkin J: Selective inhibition of endothelial cell proliferation by fumagillin is not due to differential expression of methionine aminopeptidases. J Cell Biochem. 2000, 77: 465-73. 10.1002/(SICI)1097-4644(20000601)77:3<465::AID-JCB11>3.3.CO;2-D.View ArticlePubMedGoogle Scholar
- Catalano A, Romano M, Robuffo I, Strizzi L, Procopio A: Methionine aminopeptidase-2 regulates human mesothelioma cell survival: Role of Bcl-2 expression and telomerase activity. Am J Path. 2001, 159: 721-731.PubMed CentralView ArticlePubMedGoogle Scholar
- Brooks PC, Montgomery AM, Cheresh DA: Use of the 10-day-old chick embryo model for studying angiogenesis. Methods Mol Biol. 1999, 129: 257-69. 10.1385/1-59259-249-X:257.PubMedGoogle Scholar
- Zhang LH: Novel potent class of small molecules, inhibitors of tubulin polymerization and TNF-a production with anti-tumor activity. Proc Amer Assoc Cancer Res. 2003, 44: LB-22-2Google Scholar
- Englert CR, Baibakov GV, Emmert-Buck MR: Layered expression scanning: rapid molecular profiling of tumor samples. Cancer Res. 2000, 60: 1526-1530.PubMedGoogle Scholar
- Chuaqui RF, Bonner RF, Best CJ, Gillespie JW, Flaig MJ, Hewitt SM, Phillips JL, Krizman DB, Tangrea MA, Ahram M, Linehan WM, Knezevic V, Emmert-Buck MR: Post-analysis follow-up and validation of microarray experiments. Nature Genet. 2002, 32: 509-514. 10.1038/ng1034.View ArticlePubMedGoogle Scholar
- Suzuki E, Nishimatsu H, Nagata D, Satonaka H, Goto A, Omata M, Fujita T, Nagai R, Hirata Y: Constitutive activation of proto-oncogen protein p21 induces cell cycle arrest in the G1 phase in contact-inhibited vascular endothelial cells. Hypertens Res. 2002, 25: 773-778. 10.1291/hypres.25.773.View ArticlePubMedGoogle Scholar
- Griffith EC, Su Z, Turk BE, Chen S, Chang YH, Wu Z, Biemann K, Liu JO: Methionine aminopeptidase (type 2) is the common target for angiogenesis inhibitors AGM-1470 and ovalicin. Chem Biol. 1997, 4: 461-471. 10.1016/S1074-5521(97)90198-8.View ArticlePubMedGoogle Scholar
- Sin N, Meng L, Wang MQ, Wen JJ, Bornmann WG, Crews CM: The anti-angiogenic agent fumagillin covalently binds and inhibits the methionine aminopeptidase, MetAP-2. Proc Natl Acad Sci. 1997, 94: 6099-6103. 10.1073/pnas.94.12.6099.PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article: verbatim copying and redistribution of this article are permitted in all media for any purpose, provided this notice is preserved along with the article's original URL.