Annals of Surgical Oncology Cite Track
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
10.1245/ASO.2004.03.033
Annals of Surgical Oncology 11:99-104 (2004)
© 2004 Society of Surgical Oncology
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
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
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gulec, S. A.
Right arrow Articles by Woltering, E. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gulec, S. A.
Right arrow Articles by Woltering, E. A.
Related Collections
Right arrow Tumor biology

ORIGINAL ARTICLES

A New In Vitro Assay for Human Tumor Angiogenesis: Three-Dimensional Human Tumor Angiogenesis Assay

Seza A. Gulec, MD and Eugene A. Woltering, MD

From the John Wayne Cancer Institute (SAG), Santa Monica, California; Department of Surgery (EAW), Louisiana State University Health Sciences Center, New Orleans, Louisiana; The Veterans Affairs Medical Center (EAW), New Orleans, Louisiana; and The Louisiana State University, Health Sciences Center, Stanley S. Scott Cancer Center (EAW), New Orleans, Louisiana.

Correspondence: Address correspondence and reprint requests to: Eugene A. Woltering, MD, FACS, Louisiana State University Health Sciences Center, Department of Surgery, 1542 Tulane Avenue, New Orleans, LA 70112; Fax: 504-563-4633; E-mail: ewolte{at}lsuhsc.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: A human tissue–based angiogenesis assay is needed to study the biology of angiogenesis in human tumor tissue and to tailor drug selection for patients.

Methods: Fragments of tumor tissue are embedded in fibrin gels containing medium 199, endothelial growth medium, fetal bovine serum, and {epsilon}-aminocaproic acid. Tumor implants sprout angiogenic vessels that progressively grow into the fibrin matrix. The differential growth pattern of tumor cells and angiogenic vessels in the fibrin gel matrix separates the angiogenic vessels and the tumor stroma into independently observable regions (vessel and tumor compartments). The reproducibility of the assay was tested by using fresh tissue obtained from human tumor xenografts (IMR-32 [neuroblastoma], MDA-MB-231 [breast cancer], and LNCaP [prostate cancer]) grown in nude mice and from fresh surgical breast and thyroid cancer specimens.

Results: All tumor fragments studied showed angiogenic sprouting into the fibrin matrix. This created an angiogenic vessel compartment, which was separate from the tumor fragment. The capillary nature of sprouting was confirmed histologically by factor VIII immunohistochemistry. The angiogenic growth fraction was >80% in all groups studied.

Conclusions: This assay may allow functional assessment of the angiogenic potential of human tumors and simultaneous evaluation of a therapeutic agent’s antitumor and antiangiogenic effects by virtue of its dual-compartmental structure.

Key Words: In vitro • Human • Tumor • Angiogenesis • Assay


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Tumor growth beyond a diameter of 1 to 2 mm is dependent on angiogenesis. Numerous experimental studies have concluded that primary tumor growth, tumor invasion, and the development of metastasis require neovascularization.1–8 The process of tumor growth and metastasis is complex and involves interactions among transformed neoplastic cells, supportive tissue cells (e.g., fibroblasts, macrophages, and endothelial cells), and recruited circulating cells (e.g., platelets, neutrophils, monocytes, and lymphocytes). A possible mechanism responsible for tumor growth is an imbalance, or dysregulation, of stimulatory and inhibitory growth factors within tumor systems.

Defining the biologic mechanisms responsible for angiogenesis can provide valuable insights into tumor progression and the development of metastases. Individual variation in angiogenic development (the angiogenic index: AI) may offer prognostic information by providing an index of a tissue’s neovessel growth potential. There is also an unfilled need for improved methods of quantitating an angiogenic response to support the development and application of experimental antiangiogenic interventions. The ability to assess a specific patient’s tumor response to therapy with investigational agents may also assist in the discovery of new antiangiogenic agents.

We have previously demonstrated that human placental vein fragments will develop an intense angiogenic response in a three-dimensional fibrin-thrombin matrix.9–11 We hypothesized that the cut edges of angiogenic vessels contained within human tumor xenografts or in fresh surgical specimens would grow and invade into a three-dimensional matrix. Additionally, we hypothesized that these vessels would grow over time in a progressive, linear fashion.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
We have created an in vitro human tissue–based angiogenesis model that allows the outgrowth of microvessels from a three-dimensional tissue fragment implanted in a fibrin-based matrix. The fibrin matrix is supplemented by a growth medium. The differential growth pattern of tumor cells and angiogenic vessels in the fibrin gel matrix separates the angiogenic vessels and the tumor stroma into two independently observable regions of interest (vessel and tissue compartments). The angiogenic potential of a tissue can be determined by measuring the growth of microvessels into the matrix.

Preparation of the Assay Plates
Preparation of Tumor Fragments
Fresh tumors were processed immediately after harvesting. Tumor fragments 2 mm in diameter and 1 mm thick were created and immediately embedded into fibrin gels. The fibrin gels were prepared in 96-well plates by using a specific tumor-supporting medium, as described below.

Preparation of the Tissue-Supporting Medium
A serum-free growth medium consisting of a balanced salt solution, an antibiotic-antifungal solution, and an endothelial growth medium was buffered to a pH of 7.4. Specifically, 9.5 g of medium 199 (Gibco BRL, Grand Island, NY) was dissolved in 980 mL of deionized H2O. Ten milliliters of antibiotic-antimycotic solution (Gibco) containing 10,000 U of penicillin base, 10,000 U of streptomycin base, and 25 µg of amphotericin B was added. The pH was then adjusted by adding 2.2 g of NaHCO3 (EM Science, Gibbston, NJ). This was further titrated with 1 N NaOH to pH 7.4. This solution was mixed with endothelial growth medium (Gibco) in a 3:1 ratio and sterilized by passing it through a .22-µm filter. Endothelial growth medium is a commercially available serum-free medium designed for the growth and maintenance of vascular endothelial cells.

Preparation of Fibrin Matrix Components for Tumor Fragment Embedding
A procoagulation solution was prepared by dissolving fibrinogen (.12 g; Sigma, St. Louis, MO) and .2 g of {epsilon}-aminocaproic acid in 40 mL of endothelial growth medium. Human thrombin (2 µL; Sigma) was placed in the bottom of each well of a 96-well plate and allowed to evaporate until dry.

Final Assembly of the Fibrin Matrix Tumor System and Maintenance of the Well Plates
Each tumor disk was placed in the center of a thrombin-treated well. The procoagulation solution (.2 mL) was carefully layered over the tumor fragments to prevent the formation of air bubbles in the clot. Fibrin clot formation took place within 20 to 30 minutes at 37°C. A layer of tissue-supporting medium was added over the fibrin gel. The plates were kept at 37°C in a 5% CO2/95% air humidified atmosphere. Preparation of the assay is illustrated in Figs. 1 and 2Go.



View larger version (55K):
[in this window]
[in a new window]
  		FIG. 1. Preparation steps for the three-dimensional human tumor angiogenesis assay.

 


View larger version (54K):
[in this window]
[in a new window]
  		FIG. 2. (A) Culture set in a 24-well plate; (B) an angiogenic fragment in a well.

 
Confirmation and Evaluation of the Angiogenic Response
Individual wells containing tumor fragment/angiogenic vessel compartments were examined under an inverted phase microscope. Histopathologic evaluation was performed with standard techniques.

Viability Assay
Cell/tissue viability was evaluated by using a colorimetric 3-(4,5-dimethythiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (Promega, Madison, WI). This assay is based on the cellular conversion of a tetrazolium salt into a blue formazan product. The MTT assay was performed at the end of a specified time period both on the tissue fragment and on angiogenic sprouts. Viable cells convert the colorless tetrazolium salt into a blue end product.

Histopathology and Immunohistochemistry
Histopathologic assessment was performed by hematoxylin and eosin staining. Immunohistochemistry was performed with anti–factor VIII antibody.

Angiogenic Response Measures
To determine the extent of neovessel growth, each well containing tumor fragment/angiogenic vessel compartments was visually divided into four quadrants, and each quadrant was rated on a 0 to 4 scale for the amount (length, density, and percentage of the circumference involved in the angiogenic response) of angiogenic growth (Fig. 3). A total score of 0 to 16 was calculated for each well.



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 3. Determination of the semiquantitative angiogenic score. To determine the extent of neovessel growth, each well containing tumor fragment/angiogenic vessel compartments was visually divided into four quadrants, and each quadrant was rated on a 0 to 4 scale for the amount (length, density, and percentage of the circumference involved in the angiogenic response) of angiogenic growth. A total score of 0 to 16 was calculated for each well. (A) Example of an angiogenic score of 4. (B) Example of an angiogenic score of 16.

 
Angiogenic Growth Fraction
The angiogenic growth fraction was defined as the percentage of tumor fragments that developed capillary growth into the fibrin matrix. Angiogenic fraction (AF) was the number of wells in which any angiogenic activity was observed during the observation period (day 2 to day 14) divided by total number of wells.

Angiogenic Index
The AI was the mean of the angiogenic scores (AS) of all angiogenic wells on day 14:Go


(1)

Tumor Fragment Sources
Human Tumor Xenografts
Three different human carcinoma cell lines obtained from the American Tissue Culture Collection (Rockville, MD) were used to create xenografts. The human breast carcinoma cell line MDA-MB-231 was maintained in Lebowitz’s L-15 medium (Life Technologies, Inc., Grand Island, NY) and supplemented with 10% fetal bovine serum (FBS; Life Technologies). The human neuroblastoma cell line IMR-32 was maintained in minimum essential medium (Life Technologies) and supplemented with 15% FBS, nonessential amino acids (Life Technologies), L-glutamine (Cellgro; Mediatech, Herdon, VA), and antibiotics. The human prostate cancer cell line LNCaP was maintained in 85% to 90% RPMI 1640 and supplemented with 10% to 15% FBS. Cells were harvested at subconfluence and resuspended in Hank’s balanced salt solution (Life Technologies).

Nude mice were injected with 1.5 x 107 tumor cells subcutaneously in both flank regions. Injected mice invariably grew solid tumors over a period of 4 to 6 weeks. Tumors were allowed to reach a size of 1.5 to 2 cm. Tumor harvesting was performed with sterile techniques under inhaled anesthesia with methoxyflurane, and the animals were killed immediately after tumor collection. All animal experiments were performed with the approval of the Louisiana State University Health Sciences Center’s Institutional Animal Care and Use Committee.

Fresh Human Tumor Tissues
Fresh discarded tissue samples were anonymously obtained (with Louisiana State University Health Sciences Center Institutional Review Board approval) from fresh surgical specimens of patients with breast cancer and thyroid cancer.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
All tumor types studied showed angiogenic sprouting into the three-dimensional fibrin matrix (Fig. 4). Viability of both the tumor fragments and the angiogenic sprouts was confirmed by positive colorimetry with MTT assay on day 14. Immunohistochemistry for factor VIII confirmed that the sprouts represent vascular endothelium (Fig. 5).



View larger version (113K):
[in this window]
[in a new window]
  		FIG. 4. Full angiogenic growth on day 14 in the culture. This quadrant scored 4 of 4 in determination of the angiogenic index.

 


View larger version (126K):
[in this window]
[in a new window]
  		FIG. 5. Factor VIII immunohistochemistry. (A) Positive staining of an early endothelial outgrowth in the fibrin matrix (HM). (B) Positive staining at the tumor fragment/fibrin matrix interface (HM). (C) Positive staining of the vessels inside the tumor fragment, in the fibrin matrix, and at the tumor fragment/fibrin matrix interface (LM). (D) Positive staining of the vessels inside the tumor fragment (HM). (E) Positive staining of the vessels inside the fibrin matrix (HM). LM, 40x; HM, 100x.

 
The angiogenic growth fraction (the percentage of tumor fragments that developed capillary growth into the fibrin matrix) was 77 (80%) of 96 LNCaP prostate cancer xenografts, 70 (97%) of 72 MDA breast cancer xenografts, and 69 (97%) of 72 IMR-32 neuroblastoma xenografts. In the specimens obtained from two cancer patients, the angiogenic growth fraction was 66 (92%) of 72 in thyroid cancer fragments and 67 (93%) of 72 in breast cancer fragments. The angiogenic indices (0–16 score) of the prostate, breast, and neuroblastoma tumor fragments were 13.2 ± 3.0, 12.4 ± 3.9, and 11.9 ± 3.3, respectively. The angiogenic indices for the thyroid and breast carcinoma fresh human tumor specimens were 14.2 ± 4.1 and 13.9 ± 4.9, respectively.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
There is a significant need for a reproducible human tissue–based angiogenesis assay that accurately evaluates the response of an individual tumor and its angiogenic vessels to antineoplastics, biologic response modifiers, hormones, and antiangiogenic agents. The rabbit corneal model and the chicken chorioallantoic membrane assays developed by Folkman are the two most commonly used models in angiogenesis research.12–14 Both of these models require the implantation of an angiogenic stimulator. These animal assays suffer from the need to artificially stimulate angiogenesis; therefore, the angiogenic response may not represent a true physiologic response. An ideal model should be easy to set up, should be reproducible, and should use human tissue as the source of the angiogenic response. In addition, maintaining the three-dimensional architecture of this human tissue is critical for understanding the interactions between vascular elements and supporting tissues, such as stromal and neural elements. Human umbilical vein endothelial cells have been widely used as a target for antiangiogenic agents. These cells, although human in derivation, do not allow observations on the effect of a therapy on the angiogenic switch. The angiogenic switch (the conversion of resting, or quiescent, endothelium to its proliferative phenotype) cannot be studied in human umbilical vein endothelial cell models, which by definition use proliferating monolayers of endothelial cells.

One reproducible human tissue–based angiogenesis assay is the human placental vein angiogenesis model developed by Brown et al.15 This assay is based on implantation of disks of human placental blood vessels into a fibrin gel matrix. Sprouting microvessels arise from the transected edges of the parent vessel. This response does not require the addition of stimulation with specific angiogenic factors but is serum dependent. This human tissue–based assay allows observations on the angiogenic switch from resting to proliferative phenotypes. However, the use of generic placental veins provides no information on the angiogenic response of a specific tumor-bearing patient. The human placental vein angiogenesis model is reproducible and has been used to study the human angiogenic response and to evaluate the efficacy of several antiangiogenic agents.9–11 The major drawback of vascular explant–based assays is the absence of an oncobiologic environment.

We have developed an in vitro human tissue–based angiogenesis model that allows the outgrowth of microvessels from a three-dimensional tissue fragment into a fibrin matrix. This explants are fed by a growth medium with defined growth factors. This system could be used to assay the angiogenic potential of an individual patient’s tumor. The angiogenic potential of a specific tumor can be determined by measuring the growth of microvessels into the matrix and by measuring the individual tumor’s propensity to sprout. The latter observation may lead to the development of a functional AI.

This novel system has several unique and surprising characteristics. Intact tissue architecture is maintained, including supportive stromal elements (e.g., fibroblasts), neural tissues, and endothelial tissues. The inclusion of such elements is important, because the presence of these tissues and the supporting fibrin matrix provides the framework required for angiogenesis and the growth of tumors or other tissues. The vessel growth rate typically exceeds the growth rate of the target tumor or other tissue. Thus, the higher growth rate of angiogenic vessels may be measured without interference from tissue growth. The ability to independently and accurately measure the growth of angiogenic vessels is particularly important because no known human tumor–based model has provided this important capability. The differential growth pattern of tissue cells and angiogenic vessels in a fibrin gel matrix separates the angiogenic vessels and the tissue stroma into independently observable regions of interest (vessel and tissue compartments). The compartmental structure of this novel system allows the measurement of differential effects of various antitumor or tissue-stimulatory therapies on both the tissue and angiogenic vessel components of this assay.

This model may be used to measure the angiogenic response in a variety of solid tumors. Measurements may be made in a semiquantitative or quantitative manner by direct visual examination, including computer-assisted digital image analysis. The human tissue–based system may be used to test the effects of a variety of agents on angiogenesis. Examples of such agents include growth factors, growth factor inhibitors, serum (including autologous serum), chemotherapeutic agents, external beam radiation, and in situ radiotherapy (such as those delivered via radiopharmaceutical targeting compounds). Radiolabeled peptides, monoclonal antibodies, growth factors, growth factor inhibitors, and corticosteroid hormones may also be used in this model. Chemotherapeutic agents may be studied in this model to determine the response of an individual patient’s tumor to a specific drug, drug dose, or dosing schedule.

Gulec et al.16 used this system to test for the cytotoxic and angiocidal effects of a radiolabeled somatostatin analog, 111IN-DTPA-JIC 2DL, in human tumor–nude mouse xenograft fragments. In that study, the radiolabeled somatostatin analog destroyed tumor fragments whose cells expressed somatostatin receptor subtype 2 (sst 2) but had no effect on tumor fragments whose cells lacked this receptor. Conversely, the sst 2–expressing angiogenic vessels from both sst 2–expressing and sst 2–nonexpressing tumors were destroyed by the treatment with this radiolabeled compound.

Woltering et al.17 also used this human tissue–based system to test the angiocidal effect of epothilone B treatment in fresh human tumors and normal tissues. In that study, eight human tumors and four normal tissues were treated with either nutrient medium alone or drug-containing medium beginning on the first day in culture. The study demonstrated that at doses of 10-6 M and 10-8 M, treatment with epothilone B decreased the number of wells that developed an invasive angiogenic response and limited the development of vessels that invaded the matrix. At these doses, epothilone B also caused regression of neovessels in wells that had been allowed to develop an angiogenic response.

The model may be used to provide a functional (as opposed to histological) AI. A high functional AI may help to reveal tumors with a poor prognosis even though they have a low histological AI. A disparity between functional and histological angiogenic indices may occur if circulating antiangiogenic substances (such as angiostatin or endostatin) mask the angiogenic potential of a tumor. Culturing tumors in a serum-free environment may support the unmasking of angiogenic suppressors or stimulators, thus elucidating their true angiogenic potential. The ability to test an individual tumor against a wide range of antiangiogenic agents or against a single antiangiogenic compound over a wide range of clinically relevant doses may provide clinically useful information in the future.


   FOOTNOTES
 
We describe a novel three-dimensional human tumor angiogenesis assay. This assay is unique for its three-dimensional and dual-compartmental structure. Both the tumor and angiogenesis compartments of individual human tumors can be assessed simultaneously with this assay.

Received for publication March 7, 2003. Accepted for publication September 2, 2003.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Algire G, Chalkley H, Legallais F, et al. Vascular reactions of normal and malignant tumors in vivo. 1. Vascular reactions of mice to wounds and to normal and neoplastic transplants. J Natl Cancer Inst 1945; 6: 73–85.
  2. Tannock I. The relation between cell proliferation and the vascular system in a transplanted mouse mammary tumor. Br J Cancer 1968; 22: 258–73.[Medline]
  3. Folkman J. Anti-angiogenesis: new concept for therapy of solid tumors. Ann Surg 1972; 175: 409–16.[Medline]
  4. Brown R, Weiss J. Neovascularization and its role in the osteoarthritic process. Ann Rheum Dis 1988; 47: 881–5.[Abstract/Free Full Text]
  5. Waltman D, Gitter K, Yannuzzi L, et al. Choroidal neovascularization associated with choroidal nevi. Am J Ophthalmol 1978; 85: 704–10.[Medline]
  6. Gartner S, Henkind P. Neovascularization of the iris (rubeosis iridis). Surv Ophthalmol 1978; 22: 291–312.[CrossRef][Medline]
  7. Moses M, Langer R. Inhibitors of angiogenesis. Biotechnology 1991; 9: 630–4.[CrossRef][Medline]
  8. Hobson B, Denekamp J. Endothelial proliferation in tumors and normal tissues: continuous labeling studies. Br J Cancer 1984; 49: 405–13.[Medline]
  9. Watson JC, Redmann JG, Meyers MO, et al. Breast cancer increases initiation of angiogenesis without accelerating neovessel growth rate. Surgery 1997; 122: 508–14.[Medline]
  10. Gulec SA, Gaffga CM, Anthony CT, et al. Antiangiogenic treatment with somatostatin receptor-mediated in situ radiation. Am Surg 2001; 67: 1068–71.[Medline]
  11. Jung SP, Siegrist B, Wade MR, et al. Inhibition of human angiogenesis with heparin and hydrocortisone. Angiogenesis 2001; 4: 175–86.[Medline]
  12. Gimbrone MA Jr, Cotran RS, Leapman SB, et al. Tumor growth and neovascularization: an experimental model using rabbit cornea. J Natl Cancer Inst 1974; 52: 413–27.
  13. Ausprunk DH, Knighton DR, Folkman J. Vascularization of normal and neoplastic tissues grafted to the chick chorioallantois. Role of host and preexisting graft blood vessels. Am J Pathol 1975; 79: 597–628.[Abstract]
  14. Woltering EA, Barrie R, O’Dorisio TM, et al. Somatostatin analogs inhibit angiogenesis in the chick chorioallantoic membrane. J Surg Res 1991; 50: 245–51.[CrossRef][Medline]
  15. Brown KJ, Maynes SF, Bezos A, et al. A novel in vitro assay for human angiogenesis. Lab Invest 1996; 75: 539–55.[Medline]
  16. Gulec SA, Drouant GJ, Fuselier J, et al. Antitumor and antiangiogenic effects of somatostatin receptor targeted in situ radiation with 111In-DTPA-JIC 2DL. J Surg Res 2001; 97: 131–7.[CrossRef][Medline]
  17. Woltering EA, Lewis JM, Maxwell PJ, et al. Development of a novel in vitro human tissue-based angiogenesis assay to evaluate the effect of antiangiogenic drugs. Ann Surg 2003; 237: 798–800.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
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
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gulec, S. A.
Right arrow Articles by Woltering, E. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gulec, S. A.
Right arrow Articles by Woltering, E. A.
Related Collections
Right arrow Tumor biology

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS