The protein, known as E-cadherin, had caught the eye of developmental biologists such as Francois Jacob and Gerald Edelman because it enables the early embryo to keep its shape, Birchmeier recalls. His group tested whether E-cadherin (previously known as uvomorulin) altered the stickiness of dog kidney epithelial cells. In culture, the cube-shaped cells normally pack closely together and resemble a cobblestone street. Adding an antibody that latches onto E-cadherin spurred the cells to break up and take on a ragged, irregular shape (Behrens et al., 1989). Moreover, the cells could infiltrate a layer of collagen and burrow into chunks of chick heart, signs that they had severed their moorings and become invasive. The researchers then infected normal dog kidney cells with an oncogene-carrying virus and saw the same transformation—the cells dispersed and began traveling. The notion that pro-growth oncogenes could promote cancer was familiar at the time, Birchmeier says, but “the idea that you lose something to become invasive was relatively new.” The work also demonstrated that cells could become invasive even if they weren't cancerous.
In a follow-up study, Birchmeier's group linked E-cadherin to human cancers (Frixen et al., 1991). They analyzed cell lines from many different carcinomas—epithelial tumors that are the most prevalent type of cancer in humans—and found that the invasive lines lacked E-cadherin. What's more, reinstalling the gene for E-cadherin reigned in the cells' wanderlust. Birchmeier and colleagues later discovered that an oncogene called v-src didn't alter the expression of E-cadherin. However, it did increase the number of phosphate groups stuck to E-cadherin and the bridging protein β-catenin, which connects E-cadherin to the cytoskeleton (Behrens et al., 1993). These results revealed that it's possible to reduce cell stickiness not only by reducing the amount of E-cadherin but also by altering the protein's structure.