Quantitative biological measurement in transmission electron tomography

Abstract

It has been known for some time that biological sections shrink in the transmission electron microscope from exposure to the electron beam. This phenomenon is especially important in Electron Tomography (ET). The effect on shrinkage of parameters such as embedding medium or sample type is less well understood. In addition anisotropic area shrinkage has largely been ignored. The intention of this study is to explore the shrinkage on a number of samples ranging in thickness from 200 nm to 500 nm. A protocol was developed to determine the shrinkage in area and thickness using the gold fiducials used in electron tomography. In brief: Using low dose philosophy on the section, a focus area was used prior to a separate virgin study area for a series of known exposures on a tilted sample. The shrinkage was determined by measurements on the gold beads from both sides of the section as determined by a confirmatory tomogram. It was found that the shrinkage in area (approximately to 90-95% of the original) and the thickness (approximately 65% of the original at most) agreed with pervious authors, but that a lmost all the shrinkage was in the first minute and that although the direction of the in-plane shrinkage (in x and y) was sometimes uneven the end result was consistent. It was observed, in general, that thinner samples showed more percentage shrinkage than thicker ones. In conclusion, if direct quantitative measurements are required then the protocol described should be used for all areas studied.