Membrane structure and electron microscropy the significance of physical problems and techniques (freeze etching)

Abstract

The classical methods of chemical fixation and staining are not sufficiently well understood to be very helpful in the study of the molecular structure of membranes. It was pointed out that any interpretation of electron micrographs in terms of membrane subunits must also take into account the known facts of phase contrast in the electron microscope. The possibility of misinterpretation of defocussed micrographs has been almost universally neglected in the past. Previous publications do not give sufficient information about experimental electron optical conditions to avoid being challenged on those grounds. Ideally, the focal position for a micrograph should be stated quantitatively, or focal series should be presented. Alternatively, micrographs which show the support film or the adjacent empty embedding material together with the membrane can provide some of the necessary information for an objective judgement. In view of the unknown degree of specimen preservation and the phenomenon of contrast reversal, maximum point resolution values are usually meaningless. Early hopes that the freeze-etching technique might be free of artifacts had to give way to a much more realistic appreciation of the method. The importance of a maximum cooling rate was stressed. Very little information is available as to whether the presently achievable rates are sufficient. There is much evidence for membrane splitting, but none of the arguments is completely convincing. Although the membrane particles may arise from genuine inhomogeneities, it is not yet known whether their appearance reflects their true shape and position. Methods will have to be developed to extract more information from the replica, the apparently so simple criterion of resolution being insufficient and even inapplicable under certain conditions. One of the most serious limitations of the freeze-etching technique is its lack of chemical specificity. Well-characterized and relevant model systems should be studied to learn more about the mechanisms involved in this technique which undoubtedly possesses great potential.