Diffraction patterns from stained and unstained helices: consistency or contradiction?

Addressing the issue of how best to integrate negative staining electron microscopy with X-ray fiber diffraction of macromolecular helices, we investigated staining properties of helical models using computational simulations. We compared diffraction patterns of stained and unstained representations of F-actin, myosin S1-decorated actin, and microtubules, to test the reliability of selecting well preserved specimens as those whose optical diffraction patterns most closely match the X-ray patterns of hydrated fibers. We conclude that if the stain layer is considerably thicker than the outer diameter of the specimen, this criterion is likely to be reliable. However, specimens in such thick stain layers may be resolution-limited by dynamic scattering effects such as beam-broadening. With relatively thin staining (the more practically relevant situation) "edge effects" in the stain distribution result in differences between diffraction patterns of stained and unstained specimens. Diffraction pattern changes due to molecular distortions such as shrinkage or flattening were similarly modeled. Since differences between diffraction patterns of stained and unstained helices may be due to either "edge effects" or molecular distortions, it does not appear possible a priori to distinguish between these effects. Comparison of experimental data for the (2.3 nm)-1 layer-line of tobacco mosaic virus (as model system) reveals major differences between the X-ray diffraction pattern of hydrated sols and the Fourier transform of HREM images of negatively stained specimens.