In biology, location matters. A cell's strategically placed organelles don't behave the same if they are suspended in a test tube. A gene carefully enveloped in a cell's nucleus doesn't remain functional if it's floating outside the nucleus. And a protein that normally moves long distances to do its job doesn't succeed if it's tied to one spot.
Biologists readily accept the general principle that location is intricately tied to function. Quantifying this is hard, however. There has been no well-established method to tweak a protein's location. But now, HHMI investigator Jay Groves has a technique to do just that. And the results he's getting from his experiments suggest just how important spatial organization of molecules inside cells is to human health, with implications for the immune system and cancers.
Groves and his lab members at the University of California, Berkeley, call their method "spatial mutation." While a typical protein mutation changes the overall sequence and structure of a protein, a spatial mutation changes its positioning in cells, along with its ability to move and organize.
Groves relies on a diverse lab team of biochemists, engineers, and physicists to perform these experiments. Groves did his graduate work in soft condensed matter physics—essentially, the study of materials that are soft and squishy, such as membranes. As a biophysics graduate student at Stanford University, he designed systems that allowed him to round up various membranes into discrete areas. He decided to try the technique on living biology.
"Biologists were starting to ask questions about large-scale organization that couldn't be answered by the classical tools of biology," says Groves. He suspected his method of divvying up membranes into compartments could be used to block the movement of membrane-embedded proteins. "I realized that I could get inside a cell and rearrange the organization."
He chose to study signaling between two membrane-embedded proteins, known as juxtacrine signaling. Two cells must come in close contact for this type of communication; they can't pass the chemical signal through space. Groves envisioned having one protein on a controllable synthetic membrane while its partner protein was on the membrane of a living cell.
To set up a spatial mutation, Groves and his collaborators begin with a glass slide and painted a nanometer-scale grid of metal lines on the glass using a technique (electron-beam lithography) borrowed from the computer chip industry. Then they arrange lipids—the building blocks of cellular membranes—and the embedded protein of interest within each square of the grid. Although free to move within its square of the grid, a protein can't pass the metal line, where there's a brief break in the lipids. The set-up thwarts any long-distance movement or large-scale organization of the proteins while local rearrangement and clustering are allowed to occur normally.
To test this technique, Groves first focused on T cells. Vital to the immune system, T cells recognize bits of antigen—any foreign material—displayed on the outer membrane of other cells. The T cell receptors (TCRs) that stud T cells bind to the antigens, which are displayed on molecules called major histocompatibility complexes (MHCs). This attachment triggers a major rearrangement of molecules, creating clusters of TCRs in bulls-eye patterns—what's known as an immunological synapse. The antigen-bearing MHCs rearrange on their cell surface as well. After a cascade of additional events occurs, the immune system is in defense mode.
Groves wanted to know what would happen if he blocked the rearrangement of TCRs. Would there still be signaling between the TCRs and MHCs? Would the MHCs still move? The spatial mutation's debut experiment answered no to both questions. When TCRs were confined to small areas of membrane, they could cluster locally within their individual corrals, but the full pattern of the immunological synapse could not form. The work, published in 2005, showed that the large-scale pattern of the immunological synapse affects signaling; in fact, it is part of the process of shutting off the signal.
More recently, Groves applied spatial mutations to cancer biology. There was already a suspicion that function and organization were linked in a protein called EphA2, a receptor that is pivotal in a cascade of signaling normally involved in cell migration; when the signaling is misregulated, it can lead to breast cancer.
Right away, when Groves's team watched EphA2 function on a synthetic membrane, they observed that it moved long distances after binding its ligand, ephrin-A1. This movement triggered reorganization of the cell's cytoskeleton. In tumor cells, the EphA2–ephrin-A1 complexes self-assemble into especially large clusters containing thousands of the proteins. So the scientists asked their standard question: what would happen if that movement was blocked?
They divided EphA2 into their gridded membrane before introducing a cell displaying ephrin-A1 proteins. Though the EphA2 could bind to ephrin-A1, the proteins couldn't cluster and the cell no longer rearranged its cytoskeleton. The results appeared in the March 12, 2010, issue of Science. Next, Groves wants to see small-molecule drugs tested on the system to learn what compounds mimic this disruption of EphA2 movement. Stopping the protein from forming clusters may block its cancer-causing ability, he says.
Groves thinks biology is full of other undiscovered examples of spatial organization affecting protein function. He says the technology of spatial mutations is well developed at this point, and he wants to keep expanding its applications. "If you look at a sentence, the information isn't in the number of letters of each type, it's in the order of those letters," he says. "The question in biology is: to what extent is information encoded in spatial arrangement rather than chemical content? I think it is, and to a large extent."