Dark-field Microscopy

Two approaches have been developed to circumvent the need for staining: use of image averaging of ordered arrays (pioneered by Unwin and Henderson, and not discussed further here) and dark-field imaging. The idea of dark-field microscopy is like observing stars in the night sky; dim stars are more visible in the dark background sky of the countryside rather than in the bright lights of the city. In electron microscopy, the background comes from electrons transmitted directly through the specimen as well as from electrons scattered by the substrate and surrounding structure. The STEM design removes the direct beam very efficiently and measures almost every scattered electron as it makes a flash of light upon hitting a detector. The conventional electron microscope is much less efficient for dark-field imaging and requires a dose at least ten times greater than the STEM to make an image of the same quality. Theoretically, molecules could be poised over holes to give a minimum background, but this is difficult and unreliable. A good compromise is a very thin substrate stretched across holes in a thicker film. In practice 2 nanometer (nm) thick carbon films are very easy to produce and handle, giving about 100 precent contrast for unstained DNA (also 2nm thick) on top of such a substrate. Single heavy atoms and heavy atom clusters also give high contrast in the STEM and can be used as labels.

Advantages of STEM

The high contrast in the dark field STEM images makes it easy to identify most problems in specimen preparation and outline candidate molecules. All images are acquired and stored digitally, so there is no data loss by using film for image processing. A simple computer program can measure the total mass of a particle minus the background within a contour to identify the number of subunits present and the variation from one object to another. This provides a direct link to biochemistry. Specific sites within complexes can be labeled with heavy atom clusters; much like putting pins in a map. The mass distribution shows the scaffolding while the bright spots delineate features of interest. A variety of reactivities, labels and chemistries are available.

Historical Perspective

The STEM concept was described by VonArdenne at about the same time Ruska developed the TEM (Transmission Electron Microscope) in the late 1930's. However, the STEM did not develop at that time due to lack of electronics and adequate electron sources. In the 1960's interest in the STEM was revived by Crewe and coworkers with the development of the cold field emission electron source and optimization of electronics and electron-optical components, culminating in the first visualization of single heavy atoms in the electron microscope in 1971.

This work was continued at Brookhaven in 1971 constructing STEM1, with NSF, NIH, and DOE support. STEM 1 (left) was completed in October 1977 and has operated as a User Facility continually since then, mainly with NIH and DOE Support. Significant early results include demonstration that: native and reconstituted nucleosomes had the expected mass (Woodcock et al. 1979, 1980) tri-nodular fibrinogen molecules had the mass of monomers rather than trimers (Mosesson et al. 1981) there were various size classes of intermediate filaments (Steven et al. 1982, 1983) undecagold clusters could be visualized (Safer, et al. 1982) dynein has a "bouquet" structure (Johnson et al. 1983).

STEM1 continues to operate as a User Facility with support from the NIH Biotechnology Resource Branch (grant #P41RR01777) and the US Department of Energy, Office of Health and Environmental Research (DOE-OHER). STEM2 was aquired from M. Beer (Johns Hopkins) in 1984 and used to test concepts for elemental mapping at low dose. STEM3 was designed and constructed at BNL to optimize all parameters for low-dose elemental mapping. With completion of STEM3 in June 1995, STEM2 was taken out of service. The major new features of STEM3 are: an electron energy loss spectrometer (EELS), specimen stage capable of liquid helium operation, an improved field-emission gun and an improved computer system.

References

Johnson, K.A. and Wall, J.S., Structure and molecular weight of the dynein ATPase. J. Cell Biol. 96, 669-678 (1983).   PubMed   Full Text Mosseson, M.W., Hainfeld, J., Haschmeyer, R.H. and Wall, J.S., Identification and mass analysis of human fibrinogen molecules and their domains by scanning transmission electron microscopy (STEM). J. Mol. Biol. 153, 695-718 (1981). Safer, D., Hainfeld, J., Wall, J.S. and Riordan, J.E., Biospecific labeling with undecagold: visualization of the biotin binding sites on avidin. Science 218, 290-291 (1982).   PubMed   Full Text Steven, A.C., Hainfeld, J.F., Trus, B.L., Wall, J.S. and Steinert, P.M., The distribution of mass in heteropolymer intermediate filaments assembled in vitro: STEM analysis of vimentin/desmin and bovine epidermal keratin. J. Biol. Chem. 258, 8323-8329 (1983).   PubMed   Full Text Steven, A.C., Hainfeld, J.F., Trus, B.L., Wall, J.S. and Steinert, P.M., Epidermal keratin filaments assembled in vitro have masses-per-unit-length that scale according to subunit mass: structural basis for homologous packing of subunits in intermediate filaments. J. Cell Biol. 97, 1939-1944 (1983).   PubMed   Full Text Steven, A.C., Wall, J., Hainfeld, J. and Steinert, P.M., The structure of fibroblastic intermediate filaments: analysis by scanning transmission electron microscopy. Proc. Natl. Acad. Sci. USA 79, 3101-3105 (1982).   PubMed   Full Text Woodcock, C.L.F., Frado, L.L.Y. and Wall, J.S., Direct molecular weight determination of chromatin particles by STEM. J. Cell Biol. 83, 156a (1979).   Woodcock, C.L., Frado, L.L. and Wall, J.S., Composition of native and reconstituted chromatin particles: direct mass determination by scanning transmission electron microscopy. Proc. Natl. Acad. Sci. USA 77, 4818-4822 (1980).   PubMed   Full Text

Updated 7 Aug 1995	Security Notice   Webteam   Site Map   STEM   Biology   BNL