I am studying the early steps in retroviral infection, particularly the fusion of viral and cell membranes mediated by viral envelope (Env) proteins and cellular receptors. How these proteins work is poorly understood on a molecular level. The greatest insights have come from protein crystallization studies that reveal conformational differences in Env proteins under different conditions believed to characterize different stages of the fusion process. However, such studies only provide snapshots of protein shape. Understanding how these shape changes catalyze membrane fusion will likely require ways to see details of the refolding process over time. ? New methods that detect single molecules and reveal aspects of their motion in real time are providing unparalleled insights into a variety of molecular machines like motor proteins. One of these methods, called the single molecule tether technique, involves sandwiching a molecule of interest between a surface, like a microscope slide, and a plastic bead about the size of a bacterium. The bead is about 100-1000x bigger than the molecule, just large enough to be seen as an object with some linear extent in a light microscope. Clever linkage between the molecule of interest and the bead allows motion of the molecule to be transmitted to the bead and detected. The fact that molecular motion can influence the motion of objects the size of bacteria is familiar in another context - the well-known phenomenon of Brownian motion. The latter also needs to be taken into account as a source of noise in these experiments.? ? I am trying to detect a particular kind of molecular motion, rotation, by converting it to rotation of the bead. Bead rotation can be directly observed in a microscope by attaching one or a few small fluorescent spots to the bead surface. This is done using commercially available beads coated with streptavidin and much smaller, brightly fluorescent beads coated with biotin. Biotin and streptavidin stick together like a strong glue, leading effectively to point sources of fluorescent light at various positions on a beads surface. The random Brownian rotation of the beads can be reduced by using magnetic beads, which orient in an external magnetic field like iron filings. By changing the direction of the applied magnetic field, one can control the beads orientation, which is clearly demonstrated in microscope videos. The Brownian rotation is then largely restricted to rotation about the magnetic field axis. With a collaborator in NICHD, I measured the extent of rotation about the magnetic field axis as a function of time. This is a measure of the rotational diffusion constant of the beads, and the result was consistent with a theoretical prediction based on the beads diameter and the viscosity of the surrounding liquid. While each beads orientation could be held relatively constant by the magnet, slight changes in orientation due to Brownian motion could also be detected and quantified. The magnitude of this type of Brownian motion provided a measure of the torque one could apply to a bead with the external magnetic field. Our result was roughly consistent with expectation given the magnetic field strength and magnetic specifications of the beads. These measurements give us confidence that our tracking methods are accurate and will enable us to characterize rotational motion imparted to beads by tethered molecules in the presence of Brownian noise contributed by molecules of the bathing liquid.? ? We next made a model molecular tether whose motion we believe we can predict. The model molecule is a single-stranded piece of DNA that folds into a double-stranded helical stem and a single-stranded loop, forming a lollipop configuration. DNA molecules with this configuration are used as sensitive detectors called molecular beacons because target pieces of DNA that have the right DNA sequence to bind to the loop region, force the stem to melt, which can be detected as an increase in fluorescence in appropriately labeled molecules. We put chemical groups (BrdU) on extensions of the stem region to attach the base of the DNA stem to a glass slide coated with antibodies to BrdU. We predict that melting the stem of DNA beacon molecules attached to the glass in this fashion will force the loop region to rotate about 270 degrees, since in an 8 base pair double-stranded helix (the length of the stem in our molecule) the strands wrap around each other about three quarters of a complete turn. To couple this rotation to a streptavidin bead, we put several biotin groups at the end of the loop region, and allow these beads to bind beacon DNA attached to glass. We confirmed that the beacon stem melts in this configuration when we add target pieces of DNA complementary to the loop region using a standard fluorescence beacon assay. The goal of this experiment is to convert binding of a single target DNA to 270-degree rotation of a bead tethered to the glass via a single beacon DNA. This kind of conversion of DNA binding to bead rotation has not been previously reported. If successful, with further development, it might provide a way to detect in real time other conformational changes in nucleic acid molecules triggered by different kinds of single molecule binding events, such as the binding of antigens to aptamer nucleic acid probes. This method could also provide a way to characterize unwinding pathways of regulatory RNAs.? ? We still have considerable technical hurdles to overcome. The proximity of the bead to the glass could lead to non-specific interactions that prevent rotation. Our attachment strategy could fail to anchor the beacon down at both ends, or fail to attach the loop to the bead at more than one spot, in which case the loop could rotate without turning the bead. However, we have some ideas for ways to deal with these possibilities.? ? The longer-range goal is to replace the model DNA tether with a protein molecule whose conformational change one wishes to observe, such as a viral Env molecule. Protein tethers provide considerable additional technical challenges in terms of attaching the protein molecule to the surface and the bead in a way that is rotationally constrained. For this reason, we have concentrated in the first part of this project on DNA tethers whose attachment modes are easier to engineer and whose motion is easier to predict.
Showing the most recent 10 out of 13 publications
