(Supported in part by NSF MCB9506113 to C. Mannella) The purple membrane study has been undertaken in order to ascertain whether the HVEM can collect better ED data than the lower-voltage EMs. The flatter Ewald sphere at 1200 kV gives a theoretical prediction that the high-resolution amplitudes are more accurately collected at high voltage, but this has not been demonstrated for membrane proteins, where other factors, such as crystal disorder, limit the resolution. The goals are to find the best instrument for ED of membrane proteins, use direct methods to find the phases for the diffraction data, and solve structures. The method will be tested with purple membrane and applied to VDAC. We did several experiments to learn the best way to prepare the purple membranes using both negative stain and glucose embedding. We finally able to see membranes which were a few micrometers on a side. Some of these gave six first-order spots by laser optical diffractometry, and the spots were the right spacing for the ~4 nm arrays expected for bacteriorhodopsin. Examination of the same grid with the HVEM using MRF32 film (as sensitive as LoDose film, but with a clear backgound) showed similar images, but we were unable to get laser diffraction--possibly due to the film, but also possibly due to distortion of the grid overnight. We repeated the procedure and, once again, got membranes and 2 or 4 of the expected 6 spots examining the fresh grid on the Zeiss, using SO163 film.
Showing the most recent 10 out of 252 publications
