How cells connect to each other and to the extracellular matrix (ECM) was a sticky issue in the early 1980s. Integrins, molecules that hook the cytoskeleton to ECM proteins such as collagen and fibronectin, hadn't been discovered, but evidence for a link between external and internal fibers was mounting. For example, Irwin Singer (1979) observed that extracellular fibronectin molecules closely approached—or possibly attached to—intracellular actin. Several researchers postulated that membrane-spanning receptors made the connection. A pair of papers by post-doc Wen-Tien Chen of the University of California, San Diego, and his adviser S. Jonathan Singer bolstered the idea that cells deploy different membrane receptors to couple with different components of the matrix.
A new technique devised in Singer's lab gave the researchers a clearer look at the junction between cell and surface. They reared cells on a gelatin mat, which they could roll up like a carpet, freeze, and cut into thin slices. Staining the gap with two types of antibodies pinpointed proteins clustering on both sides of the membrane. When the researchers zoomed in on a type of contact called a focal adhesion, they saw no signs of fibronectin outside the cell, although it's a key component of some cell surface junctions (Chen and Singer, 1980). Fibronectin's absence meant that cells needed a second kind of receptor to attach to the extracellular fibers found in focal adhesions, the researchers hypothesized.
A follow-up study that included more kinds of contacts (Chen and Singer, 1982). They found that fibronectin amassed in two kinds of interactions, but not in two others. Moreover, at one type of fibronectin-rich junction, microfilaments inside the cell ran parallel to the membrane. But in another sort of interaction devoid of fibronectin, microfilaments attached to the membrane head-on, like an extension cord plugging into a wall socket. These structural differences solidified the case that cells carry different receptors for different extracellular matrix proteins, says Chen (now at the State University of New York, Stony Brook). One type fastens fibronectin to microfilaments stretching along the membrane; the other joins other extracellular proteins to microfilaments that arrive perpendicular to the membrane. Chen then teamed with Kenneth Yamada of the National Cancer Institute to