NIH Awards New $14.5 million, Five-year Grant to The Scripps Research Institute

Part of NIH Roadmap Initiative, Grant will be Directed toward Membrane Protein Production for Structural Biology

By Jason Socrates Bardi

The Scripps Research Institute announced today that it has been awarded a $14.5 million, five-year grant from The National Institute of General Medical Sciences (NIGMS), a component of the National Institutes of Health (NIH). The grant, which is titled "JCSG Center for Innovative Membrane Protein Technologies," funds structural biology research on membrane proteins—an area of immense medical potential.

Now that the human genome and the genomes of dozens of different organisms have been solved, scientists all over the world are busy analyzing this genetic information. An important part of this analysis are efforts that look at how genes are expressed as proteins. Structural biology is an important part of this research because its outcome, high-resolution three-dimensional structural information, provides insights into the correlation between protein structure and function. Some day scientists may be able to interpret and catalog the structures and functions of all the proteins in the human body, which will reveal a wealth of information on the biology of human health and disease.

"Everybody feels that structural biology needs better tools for producing certain types of proteins, particularly for membrane proteins," says Scripps Research Professor Raymond Stevens, Ph.D., who is the principal investigator on the new grant.

Stevens has spent the better part of the last decade developing high-throughput structural biology tools. He has found that solving structures has become much easier and faster in general, but there are still certain types of proteins that present technological challenges that must be overcome before they yield to high-throughput methods. These include eukaryotic proteins, proteins that are involved in interactions with other proteins, and integral membrane proteins—a mere fraction of the structures contained in the Protein Data Bank are of integral membrane proteins, despite the fact that more than a third of all proteins in the body are in the membrane.

Membrane proteins are of particular interest because of their enormous medical potential. A large proportion of all drugs on the market target membrane proteins. The goal of the new grant is the development of rapid, efficient, and dependable methods to produce membrane protein samples that scientists can use to solve structures and investigate physiological functions associated with those three-dimensional structures.

The new grant pulls together a group of scientists at Scripps Research, including several who were already involved in a thus far highly successful large project developing high-throughput technology for structural biology. This project, the Joint Center for Structural Genomics (JCSG), is also sponsored by the NIH and is led by Scripps Research Professor Ian Wilson, D.Phil. (For more information on the JCSG, see a related News&Views story.)

The grant is part of the "roadmap" that NIH Director Elias Zerhouni announced last year as a guide for medical research in the twenty-first century. A series of far-reaching initiatives designed to transform the nation's medical research capabilities and speed the movement of research discoveries from the bench to the bedside, the NIH Roadmap provides a framework of the priorities the NIH must address in order to optimize its entire research portfolio and lays out a vision for a more efficient and productive system of medical research. (For more information, go to: http://nihroadmap.nih.gov/).

"It is an extremely important grant," says Scripps Research Professor Kurt Wüthrich, Ph.D, an investigator on the grant and the recipient of the 2002 Nobel Prize in Chemistry. "It fills a very important gap in the whole proteomics field."

Other Scripps Research faculty members involved in the grant include Assistant Professor Geoffrey Chang, Associate Professor M.G. Finn, Associate Professor Peter Kuhn, Professor Wilson, Professor Mark Yeager, Assistant Professor Qinghai Zhang, and Assistant Professor Scott Lesley. Lesley is also the Director of Protein Sciences at the Genomics Institute of the Novartis Research Foundation (GNF). Chang, Finn, Wilson, and Wüthrich are all members of The Skaggs Institute for Chemical Biology at The Scripps Research Institute.

Structural Biology and Membrane Proteins

Ideally, one would like to be able to stick a molecule under a microscope, look through the eyepiece, and, to paraphrase the late physicist Richard Feynman, just look at the thing.

A lot of biology can be understood simply by seeing what a molecule looks like, but the scale of biological molecules is extraordinarily small. All but the largest molecular complexes are considerably smaller than the wavelength of visible light, which means that scientists who want to know what an individual membrane protein "looks" like in nature have to rely on some other type of biophysical analysis.

Currently, three fundamental technologies are used for solving the three-dimensional structures of proteins and other biological molecules: x-ray crystallography, nuclear magnetic resonance (NMR), and electron microscopy. Scientists have used these techniques to solve the structures of tens of thousands of biological molecules.

Electron microscopy has been in use since the 1930s and relies on the use of magnetic lenses that bend a beam of electrons to image biological molecules and other tiny objects. Electron microscopy looks at a range of magnifications, from 60 times (the level of an ordinary light microscope) to 1,000,000 times. The final products of these electron images are representations of the biological structures at near-atomic resolutions—up to about 3 to 4 angstroms under the best of circumstances. That's not quite fine enough to resolve individual atoms in the structure, but more detailed structures can be modeled by combining electron microscopy with the higher-resolution techniques of x-ray crystallography and NMR.

X-ray crystallography is a technique in which a protein or some other molecule is carefully manipulated so that a crystal can form. This crystal, which is really just a large array of the molecules in solution, is then placed in front of a beam of x-rays, which diffract when they strike the atoms in the crystal. Based on the pattern of diffraction, scientists can reconstruct the shape of the original molecule.

NMR refers to the ability of atomic nuclei to reorient in a magnetic field when exposed to radiation of a particular "resonant" frequency in the radio band. An NMR spectrometer will scan a broad range of radio frequencies and record small but measurable induced voltages caused by the movement of atomic nuclei in the NMR as they go in and out of resonance. The spectrum is influenced by the shape of the molecule in which the resonating atoms reside, and scientists can use the NMR spectrum of a protein to determine its three-dimensional structure.

The three techniques have coexisted for a generation, and more and more, structural biologists have been moving beyond focusing on one technique or another to focusing on the problems that they can solve. The problem, to a certain extent, may determine the technique used. Smaller proteins may be most easily solved by NMR, larger proteins may require crystallography, and very large protein complexes may be done by electron microscopy or a combination of crystallography and electron microscopy.

"At the end of the day," says Stevens, himself a crystallographer by training, "you don't really care whether it is NMR, x-ray crystallography, or electron microscopy—you just want the structure."

In the last few years, all three techniques have been applied to solving membrane protein structures, as the work of a number of investigators at Scripps Research exemplifies:

• Using electron microscopy, Yeager and his colleagues solved and published the structure of the heart gap junction channel in a cover article in the journal Science in 1999. Heart gap channels are the specialized pores that join together heart muscle cells and allow these cells to depolarize at about the same time so that the heart beats as one single tissue and not as a loose collection of cells.
  
  
• Chang has had success using x-ray crystallography. In 2001, he published the structure of the membrane protein in a cover article in the journal Science that provided the first detailed glimpse of a membrane transporter protein—a type of large protein that sits in the cell membrane and moves other molecules in and out. Cancer cells resist chemotherapy by using these transporters, and bacterial cells use them to resist antibiotics.
  
  
• In 2002, the Stevens laboratory, in collaboration with Benjamin Cravatt's laboratory at Scripps Research, published in Science the crystal structure of the integral membrane protein fatty acid amide hydrolase at 2.8A resolution.  Fatty acid amide hydrolase degrades endocannabinoids, molecules involved in pain signaling, and is a current drug target of several pharmaceuticals companies.
  
  
• Wüthrich and his colleagues applied NMR to membrane proteins suspended in large soap-like micelles, which mimic the membrane and preserve the fold of the proteins. "You can't do conventional NMR on such things," says Wüthrich. Luckily, he and his colleagues have also been working on a technique called transverse relaxation-optimized spectroscopy (TROSY) that enables NMR to be applied to larger molecular structures.

In general, though, solving membrane protein structures has been successful only where scientists have overcome the technical challenge of producing and purifying the membranes.

The Trouble with Membrane Proteins

The main problem with using conventional techniques in structural biology to solve membrane proteins is simple and twofold—membrane proteins are hard to produce and are usually unstable once they are produced.

The easy solution to the problem is to do what structural biologists have done for decades—ignore membrane proteins and work on something else.

Scientists have gotten around the problem for years with tricks like lopping off the portion of the protein that is on the outside of the cell membrane and solving it as one would any soluble protein. But the problem with this is that it ignores what happens inside the membrane—where some of the most important biological action is.

The difficulty with solving membrane proteins starts with obtaining them. A crystallographer might need several milligrams of protein, but membrane proteins do not express well naturally. They only make up a small part of the membrane, and the membrane is only one small component of the entire cell.

In fact, says Stevens, "When you express membrane proteins at high levels, they are extremely toxic to the cell."

Even assuming success in producing sufficient quantities, many of these proteins are only stable when they are in the membrane, so they must be purified in the presence of detergents or lipids that act like a membrane, surrounding the proteins' membrane-spanning regions with detergent molecules. The conditions under which this has worked in the past are highly sensitive to such variables as detergent type, concentration, pH, and ionic strength. And these conditions are also almost always unique, demanding a lengthy trial and error search for those exact conditions in which the protein-detergent complexes will be soluble and stable. In addition, the costs of the detergents alone can easily add up to hundreds of thousands of dollars.

"The preparation of membrane proteins in the quantities and qualities needed to do structural studies is an unsolved problem," says Wüthrich. "And we hope to solve it."

Addressing the Problem

The new grant will address these central problems of membrane production and membrane stability. The Scripps Research team will examine several approaches.

On the production side, they will look at improved expression of membrane proteins by using insect or mammalian cell systems. "We know that membrane proteins will express in mammalian cells—we've done that already," says Stevens.

Yeager will be focusing on the expression of membrane proteins in mammalian and other types of eukaryotic cell lines. Eukaryotic cell lines have the advantage over bacterial cells in that they have all the enzymatic machinery in place that allows the membrane proteins to fold and assemble properly.

But, because these technologies can be expensive and time-consuming, the team is also looking at a technology that Chang and others have been developing called in vitro expression. In vitro expression does not use a cell at all, but basically takes all the components of a cell that make up the protein expression machinery—such as the genes, amino acid building blocks, and enzymes—and puts them all together in a test tube under conditions that favor expression.

In vitro expression has been used to make membrane proteins in a few limited cases, but it is far from routine. Chang calls the technique high risk and high impact. "If it works, it could revolutionize the field," he says.

In addition to addressing the problem of protein production, the grant will support a number of approaches to addressing the problem of protein stability.

One approach will be to build on the work that has been done using other proteins to help stabilize the membrane proteins. While this technique has been used before, there has never been a generalized approach to creating such a scaffold. As part of the grant, the team plans to synthesize and test a matrix of new molecules to stabilize integral membrane proteins.

Finn and Zhang will be synthesizing new lipids and detergents and helping to determine the right sorts of lipid/detergent/protein mixtures for membrane proteins. Eventually, they will scale up to several hundred different lipids and detergents, from which the best will be selected. Large quantities of these will then be produced, says Finn.

In the end, scaling up is one of the primary goals—in other words, developing the tools that will allow membrane proteins to be produced so that membrane protein structures can be solved rapidly.

"Ten years ago, people thought it would be very difficult to solve kinase structures. Now they are being solved left and right," says Stevens. "The same could happen with membrane proteins—we just need that technological breakthrough to unleash it."

Send comments to: jasonb@scripps.edu

"Everybody feels that structural biology needs better tools for producing certain types of proteins, particularly for membrane proteins," says Scripps Research Professor Raymond Stevens, who is the principal investigator on the new NIH grant.

Scripps Research Professor Ian Wilson leads a related project, the Joint Center for Structural Genomics, also sponsored by the NIH.

"[The new grant] fills a very important gap in the whole proteomics field," says Scripps Research Professor and Nobel laureate Kurt Wüthrich.

Professor Mark Yeager and his colleagues solved and published the structure of the heart gap junction channel in a cover article in the journal Science in 1999.

Assistant Professor Geoffrey Chang published the structure of a membrane protein in a cover article in the journal Science that provided the first detailed glimpse of a membrane transporter protein.

Associate Professor Peter Kuhn's lab focuses on functional proteomics and associated high throughput biophysical approaches to study biological complexes.

As part of the new grant, Associate Professor M.G. Finn will be synthesizing new lipids and detergents and helping to determine the right sorts of lipid/detergent/protein mixtures for membrane proteins.

Assistant Professor Qinghai Zhang will also be working on lipid/detergent/protein mixtures, which they hope to scale up to several hundred different mixtures.