539 



Wilhelm Menke 



lipids and proteins would have to decrease upon extraction of the lipids. The 

 water content of chloroplasts was also determined by volumetric studies. 

 Chloroplasts contain 65-80% of water, about three-quarters of which is em- 

 bedded inte r lamella rly*^. 



We have been engaged for anumber of years - initially in collaboration 

 with Professor Kratky in Graz^ , and later on in Cologne with Dr. Kreutz - 

 in studies aimed at obtaining further information about the structure of the 

 lamellar systenn by X-ray small angle scattering. We hoped thus to be able 

 to determine the thickness of the individual layers and the period of thylakoid 

 stacking, and perhaps also to discover something about the structure of the 

 layers. 



We began by determining the thickness of the protein layers from the dif- 

 fuse X-ray small angle scattering*^ . Evaluation of the scattering curves is 

 comparatively simple only if the protein layers lie individually distributed in 

 the preparation without any preferred orientation. In order to obtain prepar- 

 ations which largely satisfy these requirements, the chloroplasts were washed 

 thoroughly with water, freeze -dried, extracted with organic solvents, washed 

 again with water, freed from rapidly sedimenting portions, and freeze -dried 

 once more. The analysis of scattering curves, which was effected by various 

 procedures, showed that a lamellar system was present in the preparations, 

 and that the protein layers are 48-2 A in thickness. It was, however, discov- 

 ered that the preparations in general contain not only individual layers but 

 also stacks consisting of a small number of protein layers. It was discovered 

 later on that the scattering curj^es display a broad maximum, corresponding 

 to a Bragg' s value of about 36 A. The only explanation of this maximum is 

 that there is a periodicity within the plane of the protein lamellae. This peri- 

 odicity manifested itself repeatedly also later on, when we had changed over 

 to examining chloroplasts in an aqueous suspension and in living cells. The 

 best coincidence between the computed and experimental scattering curves 

 was found upon assumption of a quadratic lattice consisting of particles scat- 

 tering equivalent to spheres of 31.4 A in diameter. The lattice arrangement, 

 however, does not extend over the whole protein layer but is restricted to 

 crystallites. Each crystallite probably consists of 25-30 particles. The 

 crystallites are joined loosely or else separated from each other by amor- 

 phous intermediate regions . 



It appeared natural to determine next the thickness of the non-extracted 

 thylakoid membranes in freeze-dried preparations. Analysis of the scatter- 

 ing curves showed a value of 48 A in this case as well. This surprising re- 

 sult is explained by disorganization of the lipids, which takes place during 

 drying. Films of chloroplasts dried onto glass also showed some interfer- 

 ences attributable to secondary crystallization of some components of the 

 disorganized lipid mixture^ . Lipid crystals are visible in the polarization 

 microscope under intensive illumination. 



Since loss of water obviously results in partial destruction of the structure 

 of thylakoid mernbranes, isolated chloroplasts in an aqueous suspension were 

 examined next^'. These produced usable scattering curves which, however, 

 varied to a certain degree between one preparation and another. In addition 

 scattering curves of living Chlorella cells were recorded, which in turn dif- 

 fered from the scattering curves of isolated chloroplasts (Fig. 4). It was now 

 a case of finding models for the individual scattering curves, the computed 

 scattering of which would coincide with the experimental scattering curves. 

 These models should be of such a nature, that they all derive from one basic 



