Cell Surface Molecules and Receptors Tested as Targets in Treating Colon Cancer Metastases

HOUSTON—Colon cancer researchers are studying a variety of cell surface molecules and receptors as potential targets for new treatments aimed particularly at preventing or treating metastatic disease. Lee Ellis, MD, professor of surgery and cancer biology at the University of Texas M. D. Anderson Cancer Center in Houston, discussed his group’s current work in three new areas: angiopoietins as a target for blocking tumor angiogenesis, integrins, and the insulin growth factor-1 (IGF-1) system.

Tie-2 is a tyrosine kinase receptor expressed almost exclusively on endothelial cells. Dr. Ellis said that angiopoietins 1 and 2, which bind to the Tie-2 receptor, are expressed differently in normal vs cancerous tissues and may represent a useful target for treating tumor metastases.

"In vitro studies show that angiopoietin-1 (Ang-1) is an agonist. It binds to Tie-2 and initiates signaling to stabilize endothelial cells. Angiopoietin-2 (Ang-2) binds to Tie-2 with equal affinity but induces no intracellular signaling. It is thought to be an antagonist to Ang-1. It destabilizes endothelial cells, which is necessary for the initiation of angiogenesis. Ang-2 allows the endothelial cell to separate from the basement membrane, which then releases it to respond to proliferative factors such as vascular endothelial growth factor (VEGF)," Dr. Ellis said.

Maintaining Balance

When the M. D. Anderson researchers examined more than 20 human primary colon cancer samples and 5 human liver metastases, they found that tumor tissue did not express Ang-1 but did express high levels of Ang-2. Normal mucosa expressed both Ang-1 and Ang-2. "The balance seems to be what is important, as is usual in maintaining homeostasis," Dr. Ellis said.

Ang-1 and Ang-2 are made both in blood vessel tissue and outside of blood vessels. Colonocytes in nonmalignant epithelium expressed both Ang-1 and Ang-2. Malignant colonocytes expressed only Ang-2.

Dr. Ellis’s group created an in-vivo model of this situation by transfecting Ang-1 or Ang-2 into human colon cancer cell lines. Overexpression of Ang-2 greatly increased tumor growth. Overexpression of Ang-1 produced relative tumor dormancy. Ang-1 overexpression also decreased vessel counts, while Ang-2 expression increased vessel counts.

"Colon cancers frequently express Ang-2 but infrequently express Ang-1. The imbalance of the increased activity of Ang-2 over Ang-1 may contribute to the initiation of angiogenesis in colon cancer," Dr. Ellis said. "It also plays a role in vascular permeability. Ang-1 will prevent the vascular leak syndrome, but Ang-2 will induce it."

Inhibiting Angiogenesis

Researchers are investigating inhibitors of Tie-2 as a method of inhibiting angiogenesis. "Our data as well as data in gastric cancer suggests that inhibiting Tie-2, which is similar to increasing Ang-2 activity, may actually lead to an increase in angiogenesis. We should probably be investigating strategies for upregulating Ang-1 to induce tumor dormancy or to administer Ang-1 or Ang-1 peptide agonists. For some biochemical reason it has been very difficult to develop a recombinant Ang-1 protein," Dr. Ellis said.

Integrins are transmembrane proteins that mediate cell-intracellular matrix interactions. Specific integrins are expressed on endothelial cells, the most important of which for anti-tumor purposes are alpha-v-beta-3 and alpha-v-beta-5. Dr. Ellis said that these integrins bind to the extracellular matrix. They induce intercellular signaling that enhances cell survival, cell proliferation, and cell migration. Antagonists to either of these integrins can block angiogenesis. Integrin antagonists also can block endothelial cell binding to the extracellular matrix, rendering cells susceptible to apoptosis.

Preclinical studies with an integrin antagonist showed a major reduction in liver metastases in a mouse colon cancer model, according to Dr. Ellis. Treated animals also had a decrease in tumor vessel counts and an increase in tumor apoptosis.

A second study with the integrin antagonist S-247 showed significantly increased survival in treated animals vs controls, with a significant decrease in liver metastases. Dr. Ellis said that most of the mice died of increased tumor at the splenic site of primary tumor implantation, not of metastatic disease.

"Integrin antagonists appear not to be effective against primary tumor in this model but are effective against metastases," Dr. Ellis said. The anti-integrin had no apparent toxicity and will be studied further in pancreatic and gastric cancer.

No Magic Bullet

The IGF-1 receptor is overexpressed on the majority of colon cancer cells. IGF-1 increases vascular endothelial growth factor (VEGF) levels in colon cancer cells. Dr. Ellis said that IGF binds to a tyrosine kinase receptor that then dimerizes and cross-phosphorylates. This initiates signal transduction leading to induction of VEGF, which increases tumor growth and metastasis.

"Decreasing IGF-1 receptor activity inhibits tumor growth. Such tumors have fewer blood vessels, and the vessels are not as robust as in regular tumors. VEGF expression also decreases," Dr. Ellis said. "However, inhibiting the IGF-1 receptor without blocking the insulin receptor is likely to be difficult."

Angiogenesis is a dynamic process that offers many targets for intervention. According to one proposed model, upregulation of Ang-2 leads to increased hypoxia, expression of VEGF, and angiogenesis. "Then Ang-1 increases and stabilizes these new blood vessels," Dr. Ellis said. He predicted that an effective antiangiogenesis strategy will require blocking multiple pathways rather than hoping for a single "magic bullet."