When scientists picture the miniature machines that live inside cells, they often have to settle for indirect evidence and a bit of imagination. Proteins on the nanoscale – one million times smaller than a millimeter – can’t be seen with your typical microscope, so scientists turn to electrical measurements, genetic mutations, and chemical assays to deduce a rough sketch of their target’s structure. More recently, tools such as X-ray crystallography and electron microscopes have allowed scientists to see cellular proteins. But both techniques require steps that change the natural environment of the protein, and can only offer a single photograph rather than a “movie” of its dynamic changes in shape.
So when Joanna Gemel, a research associate assistant professor in the laboratory of Eric Beyer, decided to look at the structure of cellular proteins called connexins and the channels they form, she wanted a different option. Connexins are found within the membrane of a cell in groups of 6, called connexons or hemichannels. When two cells come into contact, their connexons “dock” with each other to form a pathway between the two cells called a gap junction channel. In organs such as the heart or smooth muscle, gap junctions play an important role by facilitating the rapid passage of ions and small molecules from cell to cell.
“Gap junctions are critical for the propagation of electrical impulses in the heart. Abnormalities or mutations in them can cause a lot of problems, such as arrhythmias and atrial fibrillation,” said Gemel, author of a recent paper in The Journal of Biological Chemistry. “We decided that we would like to do something different. Since we never see channels, we asked what would be the best way to see channels and learn more about them?”
The question led them to the Center for Nanomedicine, a laboratory run by Michael Allen, a research associate assistant professor in the Department of Medicine. Allen’s tool of choice is atomic force microscopy, a technology invented in the mid-1980’s that remains useful for the visualization of the very, very small. The method, known as AFM for short, uses a strategy similar to an old record player: an extremely tiny needle (2 nanometers at its tip) moves slowly across the surface of a sample, creating a topographic map of the molecular landscape.
“AFM is really good at measuring height, the resolution in the z-axis,” Allen said. “With AFM we can look at 3-dimensional architecture, and in the z-axis the resolution is a tenth of a nanometer.”
But before tapping the potential of AFM, Gemel had to first create a stretch of membrane containing only the connexin she wanted to study, a form called connexin40 that is expressed in certain regions of the heart. Through painstaking transfection, purification, and reconstitution, Gemel produced a layer as thin as a cell membrane, swarming with connexin proteins. After imaging with the microscope, the researchers produced images (like the one posted above) that resembled dense mountain ranges viewed from an airplane, bumps floating in a dark field. Remarkably, the individual subjects and even the channel opening – narrow enough to pass individual atoms – were visible, not unlike the cartoon representation of gap junctions seen in textbooks.
With the extremely fine resolution of the AFM needle, Allen and Gemel set about measuring the heights of individual objects in their sample. Even though they knew that connexin40 was the only protein present in the membrane, their images contained particles of two different heights: some “bumps” were roughly 2.5 nanometers tall, and others were approximately 4 nanometers in height. They subsequently showed that the two different-sized bumps corresponded to whether the asymmetric hemichannels were facing inward or outward in the membrane.
“While it was something that we did not expect, it was an accomplishment to be able to monitor channels from both sides,” Gemel said.
Seeing both faces of this protein with nanoscale precision was useful. But even more exciting was being able to see these miniature portals open and shut. Gap junctions close in response to calcium, so their experiments were designed to be conducted in a calcium-free environment to observe open channels. However, the AFM images showed a mix of closed and open channels, suggesting that small amounts of calcium were still present in the sample, despite the use of “calcium-free” solutions. To test this theory, Allen and Gemel added a chemical called EDTA, which vacuums up calcium, and saw a remarkable result – a small hole appeared in almost every connexin, representing an open channel.
“It was interesting to be able to throw this chemical on there and see these things pop open,” Allen said. “But one question was is this physiologically relevant – could we close them after we opened them?”
Another experiment was arranged where before and after AFM images were gathered: first while the sample was bathed in a low concentration of calcium, then after EDTA was added to bind the calcium and prevent it from interacting with the channels. Confirming the calcium sensitivity, almost all the channels went from closed to open between the first and second condition. Allen and Gemel were even able to examine the closed and open confirmations of individual connexins – a comparison that would be very difficult with a high-resolution imaging method such as crystallography that only takes a single picture.
“We don’t have to have to do two different experiments and look at 1,000 channels under these conditions and 1,000 under those conditions,” Allen said. “We’re looking at the exact same channel, so it really is a powerful approach to studying conformational dynamics. You can get a sense that the structure swells slightly and becomes a little bit taller as it closes.”
Armed with new direct evidence of what connexin40 looks like, Gemel and the Beyer laboratory can now think of new experiments to probe the structure and function of the channel. Mutated channels known to play a role in congenital heart conditions could go under the atomic force microscope, and researchers can look for clues as to why function breaks down in people with these mutations. With the help of this tiny turntable, an entire photo album of connexins is possible.
Allen, M., Gemel, J., Beyer, E., & Lal, R. (2011). Atomic Force Microscopy of Connexin40 Gap Junction Hemichannels Reveals Calcium-dependent Three-dimensional Molecular Topography and Open-Closed Conformations of Both the Extracellular and Cytoplasmic Faces Journal of Biological Chemistry, 286 (25), 22139-22146 DOI: 10.1074/jbc.M111.240002