John Hopkins scientists uncover "tags" that force proteins to cell surface

Sep 22 2005

Johns Hopkins scientists have discovered internal "shipping labels" that allow -- and perhaps force -- hundreds if not thousands of proteins to get to the surface of cells and stay there. Two natural proteins that use one of these "tags" are the ion channel that lets heart cells contract on cue, and the docking point that allows HIV, the virus that causes AIDS, into cells, the researchers discovered.

Because proteins on the cell surface are "lock-on" sites for drugs and other molecules, as well as triggers of immune reactions, the findings, described in the Sept. 11 advance online publications of Nature Cell Biology, might revolutionize efforts in drug and vaccine development, says the Hopkins team.

"A typical step in drug development is to get cells in a dish to express the protein you want to target with drugs, and then to test thousands of molecules to see which ones interact with the protein and have the effect you want," says the study's senior author, Min Li, Ph.D., a professor of neuroscience at the High Throughput Biology Center of the Johns Hopkins' Institute for Basic Biomedical Sciences.

"But if you can't get the protein to the cell surface, you can't use this screening technique. If we can force proteins to the cell surface, we can overcome obstacles that have prevented laboratory study of some really important proteins," says Li. The application of these surface tags to force protein transportation to the cell surface is the subject of a Patent Cooperation Treaty (PCT) patent application submitted by The Johns Hopkins University.

From among 25 billion randomly created, eight-building-block-long protein bits, postdoctoral fellow Sojin Shikano uncovered 65 that forced a normal protein to leave the cell's protein-building factory and go to the cell surface. By searching sequences of known human proteins, the researchers then identified those that use variations of the most potent tag they'd found, dubbed SWTY, shorthand for the four building blocks at the end of its protein sequence -- serine, tryptophan, threonine and tyrosine, in that order.

"This particular tag and its closest relatives actually mark normal proteins for delivery to the cell surface," says Li. "In some diseases, a protein that should be on the cell surface isn't, and in the lab, sometimes it's proven impossible to get a protein to the cell surface in order to study it. The tags we've found might help us force proteins to the surface, which offers real hope for overcoming these hurdles."

Laboratory studies in which the tags might be used to force a protein of interest to the cell surface are likely to be widely used fairly quickly, but Li cautions that any potential clinical applications will require understanding exactly how the tag helps the protein's transportation.

Among the "problem proteins" are those that detect odors in the nose, and the protein that's faulty in cystic fibrosis. Being able to force these to the cell surface in laboratory dishes might enable identification of more potent scents or ways to help people who can't smell, or help uncover new strategies for treating CF.

Although many scientists would say that failure to get these proteins to the cell surface means the proteins weren't assembled properly in the cell, Li says that how and where proteins are made has a lot to do with the difficulties researchers have had.

For one, proteins are made deep inside the cell; the genetic instructions for building proteins are in the cell nucleus, and proteins are assembled in a nearby "factory" in the cell. Also, scientists have long known that proteins prefer to stay put in this factory, the endoplasmic reticulum, unless they contain specific transportation instructions, much like an internal shipping label.

To figure out what tiny sequences might label the protein for delivery to the cell surface, Shikano added randomly generated eight-building-block long tags onto one end of a particular protein. He then evaluated whether the protein ended up on the cell surface instead of remaining inside the cell. The researchers found three major classes of such tags, grouped according to similarities in their sequences of building blocks, and delved into the most potent of them.

By using a computer program developed by graduate student Brian Coblitz to probe proteins' sequences, the researchers discovered that, by fairly stringent criteria, roughly 4 percent of all human proteins contain SWTY or a very close relative. The eight-block-long tag itself is part of the so-called C-terminal end of these proteins, and its existence helps explain why some engineered proteins don't go where they're supposed to go, Li says.

"If you remove a small part of the very end of a protein, it seems unlikely to disrupt how the rest of the protein folds in a three-dimensional structure, but that's what most scientists think goes wrong if a protein doesn't go to the surface," says Li. "But now we know the problem might just be a faulty transportation signal."

Given that proteins can be thousands of building blocks long, the final eight building blocks may not seem to be very important. But Li chose this size to study in part because naturally occurring proteins were already known to use similar-size bits for recognition and signaling.

"The immune system uses ones that are seven to nine blocks long to identify viral proteins or other immune triggers," explains Li.

Also, the number of possible combinations of eight-block-long protein segments provides a "reasonable number" to sort through -- 25 billion or so -- given today's high throughput technologies. To make it even easier, Shikano developed a system that would separate the wheat from the chaff before the analysis began -- if the protein wasn't taken to the cell surface by the tag, the cell died.

"If the protein went to the cell surface, the cell was in the mix, and if the cell wasn't there to be analyzed, we knew we didn't want it anyway," says Li.

Authors on the paper are Shikano, Coblitz, Sun, and Li. The researchers were funded by the National Institute of General Medical Sciences, the National Institute of Neurological Diseases and Stroke and the American Heart Association.