Radial mass density functions of vitrified helical specimens determined by scanning transmission electron microscopy: their potential use as substitutes for equatorial data

Abstract

Using STEM dark field images, we have determined linear mass densities and radial density profiles of vitrified helical particles. The samples studied are: TMV, RNA-free helical polymers of TMV coat protein (TMV-P), Salmonella typhimurium bacterial flagellar filaments and Escherichia coli pili. The difference between the profiles obtained for TMV and TMV-P shows a maximum at a radius of about 4 nm, corresponding to the RNA in TMV. Of the peaks that are resolved in X-ray diffraction analysis we can resolve the ones for TMV at radii of approximately 4.2 and approximately 6.7 nm and a shoulder at approximately 7.8 nm. Density peaks in bacterial flagellar filaments appear at radii of approximately 4.2, approximately 6.5, approximately 8.5, and approximately 10.5 nm. Accurate mass data can be obtained if the filaments are embedded in ice layers of uniform thickness; their diameters need to be similar to that of the mass standard (TMV) when these data are measured in a comparative manner. Ice layers are often not uniform, and thickness variations are well revealed in STEM dark field. The signal-to-noise ratio and contrast for the transverse projections are lower than those measured for freeze-dried specimens: half an order and one order of magnitude, respectively. The thinnest uniformly thick ice layer still containing a single layer of particles is approximately 10-15 nm thicker than the particles. Radial mass density functions that are directly determined in STEM may have a potential use as substitutes for the unreliable equatorial data in helical reconstructions of TEM bright field images of vitrified specimens.