Three-Dimensional Low Dose Reconstruction of Periodical Aggregates

Abstract

X-ray protein crystallography has contributed so much to the progress in molecular biology that one might ask whether the use of other radiations than X-rays might facilitate the analysis at least in special cases. In fact, already quite early neutron scattering has been used [1, 2] which allows the determination of the positions of the H-atoms and which facilitates phase determination with anomalous scattering. Furthermore, experiments for the use of nuclear resonance scattering have been started [3] with the aim to utilize the very strong nuclear resonance scattering of Fe⁵⁷ for phase determination of protein crystals with large protein molecules. It is, therefore, not surprising that also the use of electron rays for the crystallography of native protein crystals has been discussed and that first experiments have been undertaken in order to get some ideas concerning the feasibility of this approach [4]. One advantage is immediately evident: Due to the extremely strong scattering of electrons — compared with the scattering of X-rays or neutrons — very small specimens can be studied. It is well-known that growing of large crystals is often a difficult problem in protein crystallography. The crystals must be extremely small, if conventional accelerating voltages (order of magnitude 100 kV) are applied and if the convenient kinematical theory ought to be sufficient for the calculation of the structure factors (like in X-ray crystallography). It can be shown that the thickness of the crystal should not exceed several ten nanometers¹). Due to the limited size of the specimen, electron microscopical means are necessary for the inspection and handling of the tiny crystals. Therefore, instruments have to be used where an electron microscope is combined with a selected area diffractometer — all modern microscopes have this facility.