Angiogenesis and the Tumor Microenvironment

Preventing Angiogenesis

Anti-angiogenic therapies have been defined as those that prevent the progression of pathogenic angiogenesis. This is in contrast to anti-vascular therapies, which are defined as those that rapidly block new blood vessels to prevent circulation. Both approaches intend to reverse the tumor progress from a malignant pro-angiogenic state to a quiescent (non-angiogenic) state.24

Angiogenic Signal Inhibition

Direct angiogenesis inhibitors prevent vascular endothelial cells from proliferating or migrating toward the tumor. This class of inhibitors may also reduce the presence of pro-angiogenic factors by protein binding or regulation of gene expression. Since VEGF is the cornerstone of angiogenesis, inhibition of VEGF mediated signaling pathways has been a focus of therapeutic development.13 Pathway inhibition may be achieved by antibodies against VEGF or VEGFR, synthetic or natural proteins that bind VEGF, and small molecule inhibitors of VEGF and other receptor tyrosine kinases. Bevacizumab (Avastin®), the monoclonal antibody against VEGF, was approved in 2004 by the U.S. F.D.A. for clinical use in treating colorectal cancer and has been shown to be effective against other cancers. Bevacizumab is currently used in combination with chemotherapeutic agents to treat colorectal and lung cancers. Receptor tyrosine kinase inhibitors, such as sorafenib, sunitinib, and SU6668, are small molecules that prevent angiogenesis by inhibiting VEGF receptors and blocking VEGF signaling. Doxorubicin (Cat. No. D1515), cisplatin (Cat. No. P4394), and other chemotherapeutic agents previously approved for use in cancer treatment have been investigated and found to repress VEGF production in vitro.13,25 TNP-470 (Cat. No. T1455), caplostatin (the water soluble copolymer of TNP-740), and thalidomide (Cat. Nos. T144, T150) are other low molecular weight compounds that inhibit angiogenesis. TNP-470 prevents VEGF expression and blocks VEGF-phosphorylation of the VEGFR2 receptor.26 Thalidomide inhibits angiogenesis mediated by VEGF and bFGF, and increases tissue oxygenation in mouse fibrosarcomas. Roxithromycin (Cat. No. R4393) and clarithromycin (Cat. No. C9742), macrolide antibiotics that have anti-angiogenic activity, alter VEGF expression, although the mechanism of action is not well-defined.27

Microtubule Inhibitors

Compounds that target microtubule synthesis and tubulin also produce an anti-angiogeneic response. Reports by Tsuzuki, et al.,28 and Escuin, et al.,29 provide evidence of a relationship between disruption of the microtubule structure and disruption of the hypoxia-induced HIF-1α pathway. Microtubule targeting agents, including 2-methoxyestradiol30 (Cat. No. M6883) and albendazole31 (Cat. No. A4678) demonstrate anti-angiogenic activity, although their mechanisms of action may differ.13

Indirect Mechanisms of Inhibition

Therapies have also been developed that use indirect mechanisms to inhibit angiogenesis by suppressing angiogenic growth factors or interfering with the signaling pathways of their receptors. This indirect inhibition targets tumor cell proteins that regulate oncogenes, thereby, shifting the balance of the angiogenic switch. For example, tyrosine kinase inhibitors initially developed to target oncogenes have been found to decrease angiogenic activity as well. Trastuzumab (Herceptin®), the antibody against HER2/neu receptor tyrosine kinase, was found to downregulate the pro-angiogenic factors TFG-α, Ang-1, and VEGF, and upregulate expression of the angiogenic inhibitor Tsp-1 by gene array.32 Other signaling pathways affect the expression of pro-angiogenic or anti-angiogenic factors, and produce an indirect effect on angiogenesis. The ras oncogene confers resistance to apoptosis and increases proliferation, but ras has also been found to upregulate VEGF expression and downregulate expression of angiogenesis inhibiting proteins. In angiogenic tumor cells, Tsp-1 expression is downregulated by PI3K via phosphorylation of cMyc. Other tumor suppressor genes, including p53, PTEN, and Smad4, have been shown to increase Tsp-1 expression.24

Targeting the Tumor Vasculature

Some approaches have focused on the dysfunctional vascular system constructed by tumors as a potential target. As discussed previously, the abnormal constructions of tumor vasculature contributes to a microenvironment that acts as a barrier to chemotherapy. If the microenvironment is normalized, therapy could be more effective.12 Vasculature normalization might also reduce metastasis and provide greater oxygen levels to the tumor. Successful radiation therapy requires oxygenation of the tissue in order to generate reactive oxygen species (ROS) that damage cellular DNA and drive tumor cells into apoptosis. Jain and co-authors postulated adjusting the balance of pro-angiogenic and anti-angiogenic factors could shift the angiogenic switch, producing a more normal state of angiogenesis.2 Jain further hypothesized the judicious application of anti-angiogenic drugs can “normalize” vasculature, making delivery of drugs and oxygen to the tumor more efficient.12 The resultant improvements in tissue oxygenation and fluid removal may contribute to changes in the microenvironment, shifting it to a less tumorigenic state. This theory was tested in experimental models, which showed both bevacizumab and the antibody against VEGFR2 remodeled abnormal capillaries to produce more normal vasculature. This “normalization” using anti-VEGF agents appears to result in a therapeutic window during which other treatment methods are more successful. In a series of experiments, therapeutic radiation was found to be more effective when used in conjunction with anti-VEGF treatment.33,34 However, this therapeutic window appears to have temporal limits. Winkler, et al., found murine gliomal tumors treated with an antibody to VEGFR2 began to revert to a more malignant vasculature eight days after initial treatment.35 Treatment with anti-VEGRF2 resulted in increased angiopoietin I expression and matrix metalloproteinase activation during this normalization window. In addition to this normalization window, VEGFR tyrosine kinase inhibitors were found to initially inhibit tumor growth, but accelerate tumor growth and metastasis with prolonged use.36,37 This response may be due to hypoxic selection for more aggressive, malignant cells, or because the therapies improve the release of tumor cells into the circulating vasculature and promote metastasis.
Aspects of this research may support the hypothesis that angiogenesis inhibitors normalize the tumor vasculature by potentially pruning immature structures, resulting in improved drug treatment. However, the test results by Winkler, et al., suggested that a VEGFR2 blockade temporarily facilitates the recruitment of pericytes to tumor vessels and did not selectively prune pericyte-poor capillaries.35 To further this dichotomy, increasing the tumor vascular system instead of reducing it was shown to increase drug delivery to the tumor in a recent article. Olive, et al., found that by inhibiting the hedgehog signaling pathway in a mouse model of pancreatic ductal adenocarcinoma, the intratumoral vascular density was increased. Pancreatic ductal adenocarcinoma is highly resistant to chemotherapy. This report found treatment with IPI-926, a hedgehog signaling inhibitor, increased the vascular density of the tumor and enhanced delivery of gemcitabine to the tumor. Interestingly, this improved capillary structure was transient in a manner similar to the normalization window, and the improved vascularization reverted after several days of treatment.19 Since Rheb activation of mTOR increases the activity of HIF-1α and VEGF, the mammalian target of rapamycin (mTOR) factor is being investigated as a means to inhibit angiogenesis. Rapamycin (Cat. No. R0395) and other mTOR inhibitors may prevent angiogenesis by interfering with the mTOR promotion of HIF-1α activity. Temsirolimus, a rapamycin derivative that targets mTOR, is in clinical trials for the treatment of renal cell carcinoma.

Inhibition of Epigenetic Modification

Epigenetic modifications in tumor endothelial cells can regulate angiogenesis by silencing tumor suppressor genes. Histone deacetylase (HDAC) and DNA methyltransferase (DNMT) inhibitors are also of current interest for preventing tumor angiogenesis. Histone deacetylation and DNA methylation in cancer cells produce epigenetic suppression that downregulates genes. Hu, et al., analyzed in situ and invasive breast cancer cells and identified DNA methylation changes in all cancer cell types as compared to normal tissue.38 Debby, et al., found that gene silencing in tumor-conditioned endothelial cells was associated with histone modification, but not DNA methylation. Nevertheless, treatment of tumor endothelial cells with either the DNMT inhibitor 5-aza-2′-deoxycytidine (Cat. No. A3656) or the HDAC inhibitor trichostatin A (Cat. No. T8552) served to reverse the gene silencing, and inhibit cell growth and tumor angiogenesis.39 Various HDAC inhibitors have been found to downregulate the expression of VEGF, HIF-1α, and bFGF, and to upregulate the expression of thrombospondin-1, neurofibromin-2, and other angiogenesis inhibiting factors.39 A combination of rapamycin with the HDAC inhibitor LBH589 was found to have a synergistic effect and produce greater inhibition of HIF-1α in human prostate PC3 cells than the individual drugs.40

NF-κB and Copper

NF-κB activity is increased in cancer, and NF-κB activation modulates the expression of over 200 genes, including genes associated with apoptosis, metastasis, and angiogenesis As a result, NF-κB is known to participate in angiogenesis signaling pathways, but its role is not yet well understood. Some research has been done studying the relationship between NF-κB modulation of angiogenesis and additional factors such as reduced endogenous copper concentration.41 The participation of copper in angiogenesis was demonstrated by using penicillamine (Cat. No. P4875), a copper chelator, to reduce copper plasma levels in rabbits and block a prostaglandin E1-induced angiogenic response.42 Ammonium tetrathiomolybdate (Cat. No. 323446), an approved drug that is also a copper chelator, was shown to affect the NF-κB pathway to inhibit angiogenesis and metastasis.43

The Tumor Microenvironment

Indirect anti-angiogenic therapy is also being applied as a means to attack the tumor microenvironment and overcome drug resistance. The tumor microenvironment contributes to drug resistance because of hypoxic and acidic conditions. Several chemotherapeutic drugs, as well as radiation therapy, use oxygen to induce DNA damage to tumor cells. Hypoxia in the microenvironment results in acidosis, as lactic acid builds up due to anaerobic glycolysis. The acidic environment inhibits the efficacy of more alkaline chemotherapeutic drugs.16 Tumors also develop drug resistance over time through genetic mutation and selection of resistant phenotypes. In an experiment that focused on endothelial cells, Boehm, et al., reported an anti-angiogenic therapy directed toward tumor-associated endothelial cells, instead of directed toward tumor cells, does not produce drug resistance. The angiogenic inhibitor protein endostatin (Cat. Nos. E8154, E8279) was used to reverse angiogenesis in animal models of Lewis lung cancer, fibrosarcoma, and melanoma. The tumors were allowed to grow to a specific size within the animals and then the animals were treated with endostatin until the tumor shrank to an undetectable size. Treatment was stopped and the tumor was allowed to regrow. After repeated cycles of treatment with endostatin, the tumors remained in a dormant state for an extended period.44
An alternative to direct or indirect inhibition of the angiogenic signaling pathways regulated by VEGF is therapeutically targeting the tumor microenvironment. Angiogenesis is prevented by interfering with the recruited endothelial cells or blocking the cellular regulatory pathways that are independent of the tumor cells. Vascular disrupting agents, including tubulin destabilizers that also inhibit HIF-1α expression, may be used to disrupt tumor blood supply by targeting endothelial cells instead of the tumor.5 Flavonoids, including DMXAA (Cat. No. D5817) and baicalein (Cat. No. 465119), induce apoptosis in endothelial cells and disrupt the tumor vasculature.5,45


Some researchers have suggested a proactive approach to preventing tumor angiogenesis by using antioxidants18 or anti-inflammatory drugs as preventatives against angiogenesis (called by some authors “angioprevention”). Antioxidants have long been of interest as mediators of oxidative stress and prevention of cancer initiated by DNA damage through reactive oxygen species. However, since angiogenesis may be regulated by free radicals, peroxide, and redox-sensitive factors, including NF-κB, redox signaling may influence the expression of pro-angiogenic proteins.41 A study of the antioxidants N-acetyl- L -cysteine (Cat. No. A7250) and epigallocatechin 3-gallate (EGCG) (Cat. Nos. E4143, E4268) found that both of the antioxidants reduced the migration and invasion of endothelial cells, and downregulated gene expression of pro-angiogenic factors that control endothelial cell activation and migration.46,47 Vitamin C (ascorbic acid, Cat. Nos. A7506, A0278) demonstrated angiostatic activity in vivo that may be due to its function as an antioxidant.48 Other “angiopreventative” compounds that deactivate the angiogenic switch inhibit the recruitment and activation of endothelial cells or interfere with pro-angiogenic signaling. Two triterpenoid compounds based on oleanolic acid demonstrated anti-angiogenic activity against an immortalized Kaposi's sarcoma cell line (KS-Imm). The triterpenoids methyl 2-cyano-3,12-dioxoolean-1,9-dien-28-oate (CDDO-Me) and 2-cyano-3,12-dioxoolean-1,9-dien-28-oic imidazolide (CDDO-Im) prevent the activation of the ERK1/2 pathway after stimulation with VEGF and block NF-κB signaling by inhibiting translocation of NF-κB to the nucleus.49 A caveat to these results is the difficulty of moving promising therapies into clinical trials. Research into angiogenesis, the tumor microenvironment, and their relationships to cancer progression has expanded into multiple directions. Transferring the theoretical research to actual clinical results, however, has been a reminder that in vitro and animal models cannot completely represent the human condition. The research points toward many interesting and novel directions that intersect with multiple other biological processes and mechanisms, but always the greatest challenge will be to identify those that can truly benefit the patient.