November 20, 2008
Building a Better Foundation for Structural Studies of Membrane Proteins
Images in biology textbooks may give the misleading impression that the cell membrane is a passive envelope that does little more than keep the cell's internal contents in place. Howard Hughes Medical Institute researcher Douglas C. Rees prefers to think of the outer membrane of human cells as a dynamic boundary that sends and receives vital information about the state of affairs inside and outside the cell.
Cell membranes contain numerous proteins that act as checkpoints to control the bustling two-way flow of nutrients, hormones, signaling molecules, and ions. Myriad proteins are embedded in the membranes of cells. Those proteins are critical gatekeepers that regulate the permeability of the membrane and help to keep the cell healthy and happy.
“Given the biological and medical importance of membrane proteins, this is a critical time to investigate new methods for membrane protein structural analysis. We're excited to start this project.”
Douglas C. Rees
“They are hugely important parts of cells,” says Rees, who is at the California Institute of Technology.
Rees, who is one of a handful of researchers receiving an HHMI Collaborative Innovation Award, will use the funds to develop a novel — and hopefully more efficient and accurate - method for solving the three-dimensional structures of membrane proteins. “We want to determine their atomic structures to understand how they function as gatekeepers, but it's been a very difficult problem,” says Rees.
It is estimated that about 30 percent of the approximately 25,000 human genes code for membrane proteins. However, researchers have only determined the three-dimensional architecture of a few hundred of those proteins. According to Rees, the proteins are notoriously difficult to work with precisely because of their location within the cell membrane.
Like all structural biologists, Rees wants to see the atomic and molecular details of the proteins he is studying. To get that level of resolution, he uses x-ray crystallography, a powerful tool for “seeing” the orientation of atoms and the distances separating them within the molecules.
Before Rees can use those tools, he and his colleagues must first crystallize the proteins of interest. Only after that step is completed, can they move on to bombarding those crystallized proteins with x-rays. Computers help capture the diffraction patterns that emerge as the x-rays scatter off the atomic lattice. By rotating the crystallized protein complexes through multiple exposures, Rees's team gradually builds three-dimensional computer models that expose the architecture of these fascinating workhorses of the cell.
But all rarely goes according to plan, said Rees. Membrane proteins are not water-soluble, so scientists must first use detergents to extract them from the phospholipid membrane before they can crystallize the protein. Yet when membrane proteins are freed from the oily world they prefer, they behave erratically. This unruly behavior can create potential inaccuracies when scientists try to build atomic-scale pictures of the proteins.
Rees is specifically interested in membrane transporter proteins that pump molecules in and out of cells. He says that devising a set of techniques that permit researchers to obtain the high-resolution structures of membrane proteins in their native environment — and minimize the use of detergents — would be transformative for his own research and for the worldwide community of structural biologists.
He and his collaborators have high hopes for devising just such a technique. They are using funds from the new HHMI award to develop an approach that involves wrapping membrane proteins in a lipid envelope to mimic their natural surroundings. If the approach works, Rees and his colleagues think they will have a decent chance of producing good quality, highly ordered assemblies of membrane proteins that can be probed with x-rays and electrons.
The team will build on earlier studies by collaborator Michael Stowell at the University of Colorado, Boulder. Stowell studies the structure of proteins found in synapses. His research takes advantage of an interesting property of cell membrane lipids: Under the right conditions, they spontaneously assemble into “supramolecular” clusters in a laboratory dish.
Stowell has taken the lipid self-assembly system a step further. He has added membrane proteins to the lipids and coaxed them to form symmetrical, many-sided units called membrane protein polyhedra (MPP). The idea they will test is whether MPPs can be used as a stripped-down artificial membrane which they can then use to study the structure of a single protein embedded in the MPP using x-ray crystallography.
In his experiments, Stowell made MPPs containing a membrane protein called MscS - a channel that senses mechanical stresses. The results showed that the molecular surface of these MPPs was consistent with the x-ray crystal structure of the MscS protein that Rees and his Caltech colleagues had determined previously.
“This demonstrates that we can get structural information - in this case, about the molecular surface — from electron microscope analysis of MPPs,” Rees explains. “However, it has not yet been possible to get atomic structure information. Our working hypothesis is that the MPPs generated in these initial studies were not sufficiently ordered to extract high-resolution structural information. That is one of the goals of our project.”
In light of the many “clever suggestions” — not all of them viable — that scientists have come up with for overcoming the barriers in the field, Rees says there is a level of skepticism that makes the project high-risk. “Until you can really solve a problem with this method, people won't take it seriously,” he said.
Several challenges lie before them. For one, the researchers must show that they can mass-produce MPPs that are chemically stable with uniform size and shape. “If there is an Achilles heel in this project, that is it,” he concedes.
Another hurdle concerns the shape of the membrane polyhedra. They begin as relatively flat structures, which can then be fabricated with the exact curvature needed for each type of protein the researchers wish to study. “It's helpful to think of the membrane as a balloon,” says Rees. “It's an elastic membrane with proteins embedded in it, and we want to find out how much we need to blow up the balloon so that the proteins will be packed snugly inside.” Rees notes that this aspect of the project will benefit greatly from the expertise of collaborators Hang (Hubert) Yin at the University of Colorado and Rob Phillips at Caltech.
Yin is a protein chemist who designs and studies membrane-interacting peptides. These peptides interact naturally with the membrane bilayer in a way that pulls them into a curved conformation. Yin plans to employ peptides to help tune the propensity of the lipid system to form curved surfaces.
Phillips, who studies how proteins distort the cell membrane, “has a deep understanding of the energetics of curving surfaces,” explains Rees. Phillips' knowledge will be valuable in selecting a set of additives and lipids to use in constructing the MPPs.
“Given the biological and medical importance of membrane proteins, this is a critical time to investigate new methods for membrane protein structural analysis,” said Rees. “We're excited to start this project.”