LONDON - Loss of a molecule which helps cells stick together is a key event in determining the malignant nature of cancer cells, a team of European scientists has shown. They have proved that when benign adenoma cells stop expressing the cell adhesion molecule E-cadherin, these tumor cells become malignant.

Gerhard Christofori, group leader at the Research Institute of Molecular Pathology, in Vienna, Austria, predicted this finding may make it possible to identify drugs which will control or prevent metastasis by interfering with the genetic events that precede or follow the loss of E-cadherin.

The Research Institute of Molecular Pathology is an independent research institute and a daughter company of Boehringer Ingelheim, of Ingelheim, Germany.

Christofori, together with colleagues in Vienna and at Umea University, in Umea, Sweden, report their results in a letter to Nature, March 12, 1998, titled “A causal role for E-cadherin in the transition from adenoma to carcinoma.“

Although benign tumors grow and can damage important structures in the body, they can be surgically removed so they do not recur. But malignant tumors shed cells into the bloodstream that travel to distant organs and develop into secondary tumors, a process called metastasis. This feature is what makes many cancers so difficult to treat. Preventing metastasis could therefore make a big difference to patients' chances of surviving cancer.

Theoretically, one way to do this might be to maintain cells' ability to stick together. In many human cancers, such as those of the breast, colon, lung and prostate, the cell adhesion molecule E-cadherin is lost.

Loss Of Cell Adhesion Precedes Metastasis

However, no one knew which came first, the malignant nature of the tumor or the loss of E-cadherin. This is the question Christofori and his colleagues have now answered.

They worked with a transgenic mouse model, called Rip1Tag2, which reliably develops both benign tumors and malignant carcinomas in the beta cells of the islets of Langerhans in the pancreas. In these mice, the benign tumors express E-cadherin as normal, while the molecule is absent from the carcinomas.

Christofori and his team wanted to find out if loss of E-cadherin was an essential step in the conversion from benign adenomas to malignant carcinomas. They therefore used transgenic mice, which expressed E-cadherin in pancreatic beta cells under the control of the rat insulin promoter.

Christofori said: “This arrangement made it impossible for the tumor to switch off production of this cell adhesion molecule.“ These mice were called Rip1E-cad.

The researchers then crossed the Rip1Tag2 mice with the Rip1E-cad mice. In Rip1Tag2 offspring, beta-cell tumors developed as expected.

But in “double-transgenic“ offspring, which had a dose of each transgene (Rip1Tag2 x Rip1E-cad), tumor volumes were significantly reduced. In addition, in this group, only 7.8 percent of tumors were carcinomas, compared with 26.5 percent of tumors in the Rip1Tag2 offspring. The few cancers that did develop in the double-transgenic animals had all lost expression of E-cadherin, including that from the transgene.

The next question was whether, in live mice, loss of E-cadherin could stimulate benign tumors to invade the tissues surrounding them.

Christofori and his colleagues made use of transgenic mice that manufacture a truncated form of E-cadherin, again under the control of the insulin promoter. This truncated form of the protein binds to the normal form of E-cadherin, displacing it and destroying its usual function.

When these mice were crossed with the Rip1Tag2 mice, the team found there was a “dramatic increase“ in the proportion of carcinomas in the offspring that inherited both transgenes compared with those that inherited only one.

In the Rip1Tag2 offspring, 24.5 percent of all tumors were carcinomas, while in the double-transgenic offspring, 50.6 percent of all tumors were carcinomas.

The results showed, the authors write in Nature, that “abrogation of cadherin-mediated cell adhesion is sufficient to induce early tumor invasion and metastasis.“

Christofori told BioWorld International, “We now know that knocking out E-cadherin-mediated cell adhesion is sufficient to promote tumor progression, but loss of cell adhesion should not on its own be enough for cells to migrate and metastasize. This means that there must be events downstream of E-cadherin cell adhesion which are involved in tumor progression, but we don't know the nature of these events at this point.“

One possibility, he added, is that a molecule called beta-catenin is involved. Recent research has shown that beta-catenin is part of the E-cadherin cell adhesion complex. In addition, beta-catenin is a signal transduction molecule of the Wnt signaling pathway, which is known to play a role in the development of cancers such as colon carcinoma and malignant melanoma.

“You could imagine,“ Christofori said, “that when E-cadherin is down-regulated, the cell adhesion complex falls apart, releasing beta-catenin into the cytoplasm, where it is able to participate in the Wnt signaling pathway. It will be interesting to see if the release of beta-catenin is indeed the signal for this downstream event, and this remains to be proven.“

As a potential target for therapy, E-cadherin is a “rather disappointing“ candidate, Christofori said. “Because it is lost during tumor progression, you would have to put the gene back into every cancer cell, and this of course is impossible.“

Instead, he said, the team now intends to find out what signals regulate cell adhesion in tumor cells. Also, the team wants to identify genes that are switched on and off in response to loss of E-cadherin. *