Angiopoietins and Their Role in Colon Cancer Angiogenesis

April 1, 2002

Tumor angiogenesis is a complicated process that is regulated by numerous factors simultaneously and in a coordinated fashion.

ABSTRACT: Tumor angiogenesis is a complicated process that is regulated by numerousfactors simultaneously and in a coordinated fashion. Angiogenic factors drivethe process of neovascularization, but the initiation of angiogenesis involvespriming of endothelial cells so that they will respond to mitogenic stimuli. Theangiopoietins (Ang) -1 and -2 mediate endothelial cell stability through bindingto their endothelial cell-specific receptor, Tie-2. Ang-1 leads to endothelialcell stability, and, in vivo, to an actual decrease in angiogenesis. This islikely due to its ability to enhance adhesion to the perivascular tissues. Incontrast, Ang-2 leads to endothelial cell instability and may be an initiatingfactor in angiogenesis, priming endothelial cells for mitogenic signals. Thus, anovel antiangiogenic strategy may be one that leads to enhanced endothelial cellstability, thereby protecting endothelial cells from standard angiogenicstimuli. This may ultimately lead to tumor dormancy by transforming a rapidlygrowing tumor into an indolent tumor. [ONCOLOGY 16(Suppl 3):31-35, 2002]

Angiogenesis, a process of blood vessel formationwhereby new vessels sprout and mature from the preexisting vasculature,[1] isrequired for a variety of physiologic processes, including pregnancy, woundhealing, tissue repair, and organ regeneration. Angiogenesis also contributes tothe development of pathologic conditions such as cancer progression andmetastasis, diabetic retinopathy, psoriasis, atherosclerosis, and rheumatoidarthritis. [2,3]

Angiogenesis is a complicated process that is regulated by numerous factorsthat occur simultaneously and in a coordinated fashion. The process ofangiogenesis requires that endothelial cells (ECs) detach from pericytes and theextracellular matrix (ECM), proliferate, migrate, and form capillary tubes thatconnect to other newly developed vascular tubes.[4] The result is a primitiveyet functional vascular network. Recent evidence indicates that the presence ofspecific cytokines mediates EC "stability/instability." [5] Thesefactors either protect EC from undergoing apoptosis or facilitate EC"instability" that allows the response to mitogenic factors. Theangiopoietins are a family of proteins that mediate EC stability and survival. Abetter understanding of the biologic effects of angiopoietins in the angiogenicprocess may contribute to the development of novel therapeutic strategies.

Angiopoietins and Their Tyrosine Kinase Receptor, Tie-2

The angiopoietins are a family of growth factors identified as being specificfor the vascular endothelium. The specificity of the angiopoietins for thevascular endothelium results from the restricted distribution of theangiopoietin tyrosine kinase receptor Tie-2 (also known as TEK [6]) toendothelial cells. Four different angiopoietins—Ang-1 through Ang-4—havebeen described.[7-9] The best characterized of these are Ang-1 and Ang-2. Ang-1exerts its biologic effect by binding to Tie-2, inducing phosphorylation ofTie-2.[10,11] Ang-1 also may control the ability of ECs to stabilize thestructure and modulate the function of blood vessels. In vivo analysis bytargeted gene inactivation revealed that Ang-1 recruits and sustainsperiendothelial support cells.[12] Ang-2 is antagonistic to Ang-1 and also bindsto Tie-2, but it does not typically induce phosphorylation. However, atsupraphysiologic doses, Ang-2 also may initiate EC signaling and survival.[13]Tie-1 is an orphan receptor, but its ligand has not been identified.

Several investigators have demonstrated in vitro that Ang-1 serves as asurvival factor for ECs. Kwak and associates [14] examined the effect of Ang-1on apoptosis in human umbilical vein ECs (HUVECs). Ang-1 dose-dependently inhibited apoptosis under serum-deprived conditions, withsignificant inhibition occurring with Ang-1 doses as low as 50 ng/mL.Furthermore, the addition of 20 ng/mL VEGF (a potent angiogenic factor knownalso to be an EC survival factor) to 200 ng/mL Ang-1 augmented the antiapoptoticeffects of Ang-1, suggesting that Ang-1 acted in conjunction with VEGF.

Similar results were obtained by Papapetropoulos et al,[15] who alsodemonstrated dose-dependent stabilization of HUVEC network organization byAng-1. In addition, these authors demonstrated that this response was indeeddependent on Tie-2 activation, as addition of a soluble form of Tie-2, but notTie-1, completely blocked the effects of Ang-1. It was also demonstrated thatthe signaling pathway by which Ang-1 protects ECs from apoptosis is likelythrough phosphorylation of the survival serine-threonine kinase, Akt. [16,17]This finding occurred in association with the up-regulation of the apoptosisinhibitor, survivin, in ECs and protection of endothelium from apoptosis.[16] Inaddition, transfection of a dominant-negative survivin construct abrogated theability of Ang-1 to protect cells from undergoing apoptosis. These data suggestthat the activation of antiapoptotic pathways mediated by Akt and survivin inECs may contribute to Ang-1 stabilization of vascular structures duringangiogenesis.[16]

It is likely that Ang-1 works in conjunction with VEGF to help stabilizevascular networks. Ang-1 appears to recruit periendothelial support cells, [12]and this interaction may be required for EC survival. This is supported byevidence showing that Ang-1 knockout embryos are able to undergo VEGF-dependentangiogenesis, but ECs are unable subsequently to interact with periendothelialsupport cells.[12] This deficit leads to vascular regression. Ang-2 isantagonistic to Ang-1 and thus leads to EC destabilization. In the presence ofVEGF, this destabilization leads to robust angiogenesis.[18]

Inhibition of the activity of the Angs by soluble Tie-2 has been investigatedas a method of inhibiting angiogenesis. Lin and colleagues[19,20] constructed anadenoviral vector containing the mouse extracellular domain coding region ofTie-2, which can systemically deliver recombinant soluble Tie-2 (AdExTek)capable of blocking Tie-2 activation. Administration of soluble Tie-2 (AdExTek)inhibited the growth in both primary and metastatic tumors in mice. However, itis unclear whether the observed effects are due to sequestration of Ang-1 orAng-2.

Figure 1 depicts the roles of Ang-1 and Ang-2 in relation to Tie-2 expressionand activity.

Angiopoietins in Gastrointestinal Cancer

Few studies have examined the role of angiopoietins in gastrointestinalcancer progression and metastasis. In our laboratory, we investigated whetherangiopoietins are expressed by human colon carcinoma.[21] Using reversetranscriptase-polymerase chain reaction (RT-PCR), Ang-1 and Ang-2 expressionwas measured in normal colonic mucosa, colon cancer specimens, and colon cancercell lines. Results showed a relatively equal frequency of expression of Ang-1and Ang-2 in normal colonic mucosa. However, 11 of 11 colon cancer specimensexpressed Ang-2, while only 6 of 11 (54%) expressed Ang-1 (P = .04). Similarly,the majority of colon carcinoma cell lines expressed Ang-2 (14/18) (78%),whereas fewer (7/18) (39%) expressed Ang-1 (P = .04). These preliminary studiessuggest that an imbalance of activity of Ang-2 over Ang-1 may play a role incolon cancer angiogenesis, whereas angiopoietins are expressed with relativelyequal frequency in normal tissues. We hypothesized that the balance ofangiopoietin expression in normal tissues likely promotes homeostasis, whereasthe imbalance of Ang-2 over Ang-1 in malignant tissues leads to the initiationof angiogenesis.

To examine further the role of angiopoietins in human colon cancer, we usedimmunohistochemical techniques to assess Ang-1 and Ang-2 expression in 20primary colon cancer and 5 liver metastasis specimens. Results demonstrated thatAng-1 and Ang-2 both were present in all normal colonic mucosa specimensstudied. However, Ang-1 was not expressed in any of the colon cancer specimens,while Ang-2 was expressed in all of them. To determine the cell of origin of theangiopoietins, we performed immunofluorescent double staining with antibodiesthat bind to either Ang-1 or Ang-2 and CD-31 (endothelial cell marker). Resultsdemonstrated coexpression of the angiopoietins in the vasculature of normalcolonic mucosa. However, angiopoietins also were expressed in surroundingtissues. The same staining pattern was noted for Ang-2 in both normal colonicmucosa and tumor tissue. To confirm that the angiopoietins were expressed in thesurrounding epithelium (both normal and malignant), double staining withcytokeratin-22 (CK-22, an antibody that recognizes epithelial cells) andantibodies to the angiopoietins was performed.

Results showed colocalization of Ang-1 and Ang-2 in colonic epithelial cellsin normal colonic mucosa. Ang-2 colocalized with malignant colonic (CK-22)epithelium, but, as in the above studies, there was no evidence of Ang-1expression in malignant colonic epithelium. The same staining pattern held truefor colon cancer liver metastasis specimens. The above data suggest that theimbalance of activity of Ang-2 over Ang-1 may be associated with the initiationof angiogenesis in malignant tissues.[21]

To investigate this hypothesis further, we performed in vivo studies withcell lines stably transfected with full length cDNA constructs for Ang-1 orAng-2.[22] The HT-29 human colon cancer cell line was transfected with the emptyvector (pcDNA.3) as a control of the Ang-1 and Ang-2 constructs. Stable clonesthen were injected subcutaneously in nude mice. In the initial study, miceimplanted with tumors overexpressing Ang-2 had a tremendous increase in tumorgrowth (Figure 2). Due to the large tumor volumes, all of the mice weresacrificed at one time point. Tumor weight was significantly decreased in theAng-1 group and was increased in Ang-2-overexpressing tumors. Analogous tothis finding, vessel counts were decreased in tumors grown from Ang-1-overexpressingcells and were increased in tumor cells overexpressing Ang-2 (Figure3). Thispattern held true for tumor cell proliferation as demonstrated by proliferatingcell nuclear antigen (PCNA) staining.

To determine if the decrease in vessel count in Ang-1-overexpressing tumorscould lead to a potential decrease in tumor growth as compared with controlgroups, the study was repeated with mice sacrificed at a later time point. Ang-2-overexpressingtumors again grew rapidly; all mice in this group had to be sacrificed due tolarge tumor burden. However, over a longer time period, the growth curvesbetween the control cells and Ang-1-overexpressing cells diverged; tumorgrowth was relatively dormant in Ang-1-overexpressing cells while it continuedto be exponential in control cells. These data suggest that Ang-1 overexpressionpossibly stabilized ECs, preventing their release from pericytes and thebasement membrane, thus inhibiting initiation of EC proliferation andneovascularization.

Other investigators recently reported results of similar studies in varioustumor types. Hayes and colleagues demonstrated that overexpression of Ang-1 inthe human breast cancer cell line MCF-7 decreased tumor growth threefold.[23]Furthermore, Etoh and colleagues conducted a comprehensive study of the role ofAng-2 in gastric carcinoma. [24] These authors, using 85 gastric cancerspecimens, demonstrated that high levels of tumor Ang-2 expression wereassociated with more frequent vascular involvement and more advanced diseasestages as compared with low levels of tumor Ang-2 expression. Patients whosetumors had high Ang-2 mRNA expression had poorer survival than did those withrelatively low Ang-2 expression.

As in our studies, these authors found that Ang-2 was expressed predominantlyin malignant tissues as compared with normal tissues. In in vivo studies,gastric cancer cells transfected with an Ang-2 construct developed highlymetastatic tumors that were hypervascular when compared with control cells.Further studies also demonstrated that the production of Ang-2 led to theupregulation of proteases such as matrix metalloproteases-1 and -9 andurokinase-type plasminogen activator in ECs in the presence of VEGF.[24]


Results from the studies described herein support the hypothesis that theimbalance of Ang-2 activity over Ang-1 activity is important in the angiogenicprocess and, in fact, may be an initiating factor in angiogenesis.[18] Incontrast to approaches that target the activity of the angiogenic tyrosinekinase receptor, a novel anti-angiogenic approach may be one that actuallyincreases the activity of the agonist Ang-1, thus potentially stabilizing ECssuch that they cannot respond to mitogenic signals in the microenvironment. Thisstrategy potentially would lead to tumor dormancy and would be unlikely to havedose-limiting toxicities. Thus, this strategy represents the paradigm ofconverting a rapidly growing tumor ("acute disease") into a slowlygrowing or dormant tumor ("chronic disease").


1. Folkman J: Tumor angiogenesis: Therapeutic implications. N Engl J Med285:1182-1186, 1971.

2. Folkman J: Angiogenesis in cancer, vascular, rheumatoid and other disease.Nature Med 1:27-31, 1995.

3. Battegay EJ: Angiogenesis: Mechanistic insights, neovascular diseases, andtherapeutic prospects. J Mol Med 73:333-346, 1995.

4. Fidler IJ, Ellis LM: The implications of angiogenesis to the biology andtherapy of cancer metastasis. Cell 79:185-188, 1994.

5. Liu W, Ahmad SA, Reinmuth N, et al: Endothelial cell survival andapoptosis in the tumor vasculature. Apoptosis 5:323-328, 2000.

6. Partanen J, Armstrong E, Makela TP, et al: A novel endothelial cell surface receptor tyrosine kinase withextracellular epidermal growth factor homology domains. Mol Cell Biol12:1698-1707, 1992.

7. Grosios K, Leek JP, Markham AF, et al: Assignment of ANGPT4, ANGPT1, andANGPT2 encoding angiopoietins 4, 1 and 2 to human chromosome bands 20p13,8q22.3tq23 and 8p23.1, respectively, by in situ hybridization and radiationhybrid mapping. Cytogenet Cell Genet 84:118-120, 1999.

8. Holash J, Wiegand SJ, Yancopoulos GD: New model of tumor angiogenesis:Dynamic balance between vessel regression and growth mediated by angiopoietinsand VEGF. Oncogene 18:5356-5362, 1999.

9. Valenzuela DM, Griffiths JA, Rojas J, et al: Angiopoietins 3 and 4: Diverging gene counterparts in mice and humans.Proc Natl Acad Sci USA 96:1904-1909, 1999.

10. Jones N, Master Z, Jones J, et al: Identification of Tek/Tie2 bindingpartners. Binding to a multifunctional docking site mediates cell survival andmigration. J Biol Chem 274:30896-30905, 1999.

11. Fujikawa K, de Aos Scherpenseel I, Jain SK, et al: Role of PI 3-kinase inangiopoietin-1-mediated migration and attachment-dependent survival ofendothelial cells. [erratum appears in Exp Cell Res Feb 25;255(1):133, 2000].Exp Cell Res 253:663-672, 1999.

12. Suri C, Jones PF, Patan S, et al: Requisite role of angiopoietin-1, aligand for the TIE2 receptor, during embryonic angiogenesis [see comments]. Cell87:1171-1180, 1996.

13. Kim I, Kim JH, Moon SO, et al: Angiopoietin-2 at high concentration canenhance endothelial cell survival through the phosphatidylinositol 3'-kinase/Aktsignal transduction pathway. Oncogene 19:4549-4552, 2000.

14. Kwak HJ, So J-N, Lee SJ, et al: Angiopoietin-1 is an apoptosis survivalfactor for endothelial cells. FEBS Lett 448:249-253, 1999.

15. Papapetropoulos A, Garcia-Cardena G, Dengler TJ, et al: Direct actions ofangiopoietin-1 on human endothelium: Evidence for network stabilization, cellsurvival, and interaction with other angiogenic growth factors. Lab Invest79:213-223, 1999.

16. Papapetropoulos A, Fulton D, Mahboubi K, et al: Angiopoietin-1 inhibitsendothelial cell apoptosis via the Akt/survivin pathway. J Biol Chem 275:9102-9105, 2000.

17. Kim I, Kim HG, So JN, et al: Angiopoietin-1 regulates endothelial cellsurvival through the phosphatidylinositol 3'-kinase/Akt signal transduction 18.Holash J, Maisonpierre PC, Compton D, et al: Vessel cooption, regression, andgrowth in tumors mediated by angiopoietins and VEGF. Science 284:1994-1998,1999.

19. Lin P, Polverini P, Dewhirst M, et al: Inhibition of tumor angiogenesisusing a soluble receptor establishes a role for Tie2 in pathologic vasculargrowth. J Clin Invest 100:2072-2078, 1997.

20. Lin P, Buxton JA, Acheson A, et al: Antiangiogenic gene therapy targetingthe endothelium-specific receptor tyrosine kinase Tie2. Proc Natl Acad Sci USA95:8829-8834, 1998.

21. Ahmad SA, Liu W, Jung YD, et al: Expression of angiopoietin-1 and -2 inhuman colon cancer: Possible mechanism for initiation of angiogenesis. Cancer92:1138-1143, 2001

22. Ahmad SA, Liu W, Jung YD, et al: The effects of angiopoietin-1 and -2 ontumor growth and angiogenesis in human colon cancer. Cancer Res 61:1255-1259,2001.

23. Hayes AJ, Huang WQ, Yu J, et al: Expression and function ofangiopoietin-1 in breast cancer. Br J Cancer 83:1154-1160, 2000.

24. Etoh T, Inoue H, Tanaka S, et al: Angiopoietin-2 is related to tumorangiogenesis in gastric carcinoma: Possible in vivo regulation via induction ofproteases. Cancer Res 61:2145-2153, 2001.