Home > Research > Section Shrinkage

Collapse of plastic sections in the electron microscope (Luther et al, 1988)

    The traditional method of 3D reconstruction of tissues and cellular components has been serial-section reconstruction in which the same region is imaged along a ribbon of sections cut from a plastic-embedded block of the sample.  The depth-resolution of this method is limited by the thickness of the sections to ~30-50 nm.  Much higher depth resolution, ~4-7 nm, can be obtained by electron tomography [1-3].  In this method, many different views (~50-150) of a sample tilted about one or two axes are computationally combined.  The method requires that the original sample (i.e. the selected region) remains constant during the microscopy.  We have found, along with others [4-6], that sections shrink in depth, often drastically, immediately upon viewing in the electron microscope.  On account of its rapidity, we refer to the shrinkage as section collapse.   For plastic-embedded biological sections, we have so far found that the shrinkage cannot be eliminated but it can be reduced using low-dose methods and cryo methods [4,6].  Because the collapse is normal to the surface of the section, the untilted view, as in routine 2D microscopy, is unaffected (bar a small amount of planar shrinkage, see later).  Serial section reconstruction is also unaffected by section collapse, but its poor depth resolution compels the use of electron tomography.  In the reviews (7,8) the characteristics of the collapse are described and how its effects are taken into account for electron tomography.


Simple illustration of section shrinkage (1).  (a) A typical section of muscle, represented as a slab, showing edge-on views of the hexagonal myo-filament lattice.  The direction O, which is along the electron beam gives the initial untilted view of the section.  If the section is tilted about the myofibre axis (arrow), by +30o or 30o, then either of the directions A or B will be vertical, and we obtain a projection of a principal lattice view.   (b) shows the same sample collapsed to 70% of the original thickness.  Now the tilt required for the same lattice projections is about 40o.


Simple illustration of section shrinkage (2).  Take for example, a cylindrical object, left, viewed edge-on, divided into 10 degree segments as labelled.  Suppose the viewing window is +/- 60 degrees about the central 0 degrees view.  If the object shrinks by 50%, as shown in the right image, the same +/- 60 degrees viewing window now shows a smaller number of segments.  



Measurement of section collapse (lower trace) and planar shrinkage (upper trace) as a function of electron dose (3). 

The plots show that typically the section shrinks laterally to about 95% and collapses to 75% of its starting values after 5 min (an accumulated dose of 1300 e-/nm2).  The collapse is most rapid in the first 3 min and thereafter the section stabilises with only a little further collapse and lateral shrinkage.  After 15 minutes, the dose rate was increased nearly 9 fold to 380 e-/nm2/s.  This caused a new collapse curve taking the lateral shrinkage to 90% and collapse to 70% before again stabilising.  Since the curves level off after each set dose rate, this indicates that the changes that occur in a sample allow it to dissipate the energy at the set dose rate.  This implies that the lower the beam intensity used to collect a tilt-series, the smaller is the total shrinkage of the sample.   For this reason, it is necessary to preirradiate a section in order to induce shrinkage beyond the rapid phase.  In the levelled off region, using low-dose electron microscopy, several electron micrographs can be recorded for a complete tilt series with little further change in dimensions. 



1. Frank, J. Electron Tomography, J.  Plenum, New York, pp 39-60, 1992. 

2. Baumeister, W. et al.  Electron tomography of molecules and cells.  Trends Cell Biol. 9, 81-85, 1999

3. Koster, A.J. et al.  Perspectives of molecular and cellular electron tomography. J. Struct. Biol. 120, 276-308, 1997.

4. Luther, P.K., Lawrence, M. and Crowther, R.A.  A method for monitoring the collapse of plastic sections in the electron microscope.  Ultramicroscopy 24, 7-18, 1988.

5. Bennett, P.M. Decrease in section thickness on exposure to the electron beam; the use of tilted sections in estimating the amount of shrinkage. J. Cell Sci. 15, 693-701, 1974. 

6.  Braunfield, M.B. et al. Cryo automated electron tomography: towards high-resolution reconstructions of plastic-embedded structures.  J. Microsc. 174, 75-84, 1994.

7.  Luther, P.K.  Section shrinkage and radiation damage" In Electron Tomography, Ed: Frank, J.  Plenum, New York, pp 39-60, 1992. 

8. Luther, P.K. (2001) Section shrinkage: Impact for routine and 3D electron microscopy, Microscopy and Analysis, 7-9.