ACADEMIA

For a glycerol molecule, a measly angstrom's difference in diameter is a road-closed sign: You can't squeeze through unless you are a sleek, water-molecule-sized sports car, say scientists at the University of Illinois at Urbana-Champaign. The National Institutes of Health funded the study, which involved the use of supercomputers at the Pittsburgh Supercomputer Center and the National Center for Supercomputing Applications at Illinois. The roadway is in aquaporins, a class of proteins that form trans-membrane channels in cell walls in all forms of life. They allow for water movement between the cell and its environment. A subfamily of aquaporins allows slightly larger molecules, such as glycerol, to pass, too. In humans, 11 aquaporins have been identified, mostly in the kidney, brain and lens of the eye. Impaired function has been implicated in a variety of diseases. Aquaporins are a target of scrutiny for the Theoretical and Computational Biophysics Group at the Beckman Institute for Advanced Science and Technology. Using steered molecular dynamics, Beckman researchers have solved a mystery that years of protein crystallography couldn't accomplish. Reporting in the August issue of Structure, they show that the main structural difference that makes an aquaporin a glycerol channel is a channel that is just a hundred-millionth of a centimeter -- an angstrom -- wider than a normal water channel. So even if glycerol molecules line up properly, as do water molecules to pass through a pure water channel ( as documented by researchers in the same lab in 2002 ), the slightly larger sugar molecule is out of luck. The point of entry, known as a selectivity filter, is the most narrow, but there are other tight barriers blocking the way as well, said Emad Tajkhorshid, assistant director of research in the Beckman lab. "Membrane proteins are difficult to crystallize," he said. "We don't have the known structure of many of them. There has been a lot of recent progress, and for aquaporins we've got four structures available, which is really exceptional for membrane channels." For the new study, his team focused on two of them. "Both were from the same bug, E-coli. One was a pure water channel. The other is a glycerol channel," Tajkhorshid said. "Structurally they are similar. Researchers have tried to convert a water channel to a glycerol channel, or the other way around, by mutating amino acids that line the pore of the channel, but they have failed." The E-coli proteins studied were AqpZ, a water channel, and GlpF, a glycerol channel. Side-by-side in computer-generated images the channels appear virtually identical. The Beckman teamed pushed glycerol through the channels, calculated the energetics and looked for barriers. "Nature is using a very, very simple idea here," Tajkhorshid said. "Just by making a channel narrower, only water is allowed to pass through the pure water channel; by making it a little bit bigger in the other channel both glycerol, as well as longer, linear sugar molecules, and water can permeate the channel." While channel sizes had appeared slightly different after crystallizing the proteins in the past, researchers believed the channels could be manipulated by inducing the surrounding amino acids to create a hydrophobic or semi-hydrophobic lining required for glycerol passage. Success in doing so could have created new targets for drug therapies. However, it turns out, the amino acids are the same around both channels, Tajkhorshid said. So his team now is looking beyond the amino acids directly lining the channel to find what it is that forces changes in size. The same principles, he added, likely apply to all selective protein channels. Understanding the principles could provide new, effective pharmaceutical targets to control the channels to help treat disease. Study co-authors were group director Klaus Schulten and Yi Wang, a doctoral student in molecular and cell biology.