The Structure of CF0

CF0 consists offour different subunits, I, II, III, and IV. Subunit IV (homology to E. coli a) was initially missed by SDS-PAGE as a component of CF0 [62] because it stains only weakly by Coomassie Blue. Later, subunit IV was identified in silver stained gels as a genuine component of CF0F1 [61].

Subunit III (homology to E. coli c) exists in multiple copies and is the major component of CF0. It forms a stable complex that remains intact even during SDS gel electrophoresis and requires heating in SDS-buffer for dissociation into monomers -55] - The subunit-III complex has an oblate shape with a membrane spanning length of 6.1nm and a diameter of 6.2nm and a stoichiometry of III12 [55] was suggested. Later, investigations of the subunit III-complex with atomic force microscopy (AFM) allowed the identification of the individual subunits in the complex resulting in 14 monomers per complex -53] - The stoichiometry of the subunits in the ring seems to be fixed by the shape of the subunits and the contacts to their nearest neighbors as revealed by incomplete complexes that maintain their diameter [54]. Comparison of AFM images of CF0, III14IV and III14 show an additional density inside the III -ring in sub-complexes that contain subunit IV -56]. This lead to the assumption that subunit IV is located in the centre of the III-ring. However, this notion is in contradiction to the current structural and functional understanding of CF0. An alternative explanation for the presence of the additional density is the variable amounts of lipids that plug the ring similarly as observed for the c-ring of I. tartaricus [73].

For modeling the III14-ring, the c-ring of I. tartaricus (1YCE, [51]) was used as a template (Figure 9.3 top, centre). This ring consists of only 11 subunits. Each subunit is a helical hairpin. The N--erminal helices form an unusually tightly packed inner ring. The C-terminal helices fill the grooves between the N-terminal helices and form an outer ring. Sodium ions bind between C --erminal and N-terminal helices in the center of the membrane, where the diameter of the ring is smallest. Since I. tartaricus is a sodium-translocating ATP synthase, it is likely that these sodium ions identify the potential proton-binding sites in H+-ATP synthases.

With 11 subunits, the c-ring of I. tartaricus is significantly smaller in diameter than the III14-ring in chloroplasts. For modeling of the III-ring, the inter-subunit packing of the homology models of subunit III was kept similar as in I. tartaricus, but the stoichiometry of subunits and the diameter of the ring were adjusted according to the AFM data ([53, 74], Figure 9.3 top, left). The modeled ring matched the surface topology measured by AFM (Figure 9.3 top, right).

Currently, no experimental data on the structure of subunit IV (subunit a) is available. For placing the subunit IV relative to the III-ring, the model of the E.

Figure 9.3 Top: Superposition of the structure of the subunit III14-complex from atomic force microscopy (grey, left) with the homology model of subunit III14 (light brown, right), based on the structure of III14-complex from Ilyobacter tartaricus (center, 1YCE). Bottom: Superposition of the CF0-part of the three-dimensional map of CF0F1 from electron microscopy (grey) with homology models. The subunit III14-complex is shown in light brown. The N-terminal domains of subunit I and II are based on the structure of this region of the b-subunit from E. coli (1A9U, centre), blue. The model of subunit IV is based on the structure of the a-subunit from E. coli (1C17, centre) blue.

Figure 9.3 Top: Superposition of the structure of the subunit III14-complex from atomic force microscopy (grey, left) with the homology model of subunit III14 (light brown, right), based on the structure of III14-complex from Ilyobacter tartaricus (center, 1YCE). Bottom: Superposition of the CF0-part of the three-dimensional map of CF0F1 from electron microscopy (grey) with homology models. The subunit III14-complex is shown in light brown. The N-terminal domains of subunit I and II are based on the structure of this region of the b-subunit from E. coli (1A9U, centre), blue. The model of subunit IV is based on the structure of the a-subunit from E. coli (1C17, centre) blue.

coli ac12-complex served as a template, based on computational structure prediction (1C17, [ 19] : Figure 9.3: bottom, centre). The resulting III [4IV-model was placed as a whole into the CF0-part of the 3D-map of CF0F1 (Figure 9.3 bottom, left). The CF0-part in the 3D-map is considerably larger than the modeled III[4IV, which makes the placement tentative. The reason for the discrepancy in size is the presence of a detergent micelle that shields the hydrophobic regions of CF0, which are usually integrated into the thylakoid membranes. The outline of the III-ring can be recognized in the 3D-map inside the detergent micelle. The III-ring protrudes at the bottom from the detergent micelle (Figure 9.3 bottom, left) and is closed by an unidentified density, which is similar to the lipids that plug the c-ring of I. tartaricus [73]. Adjacent to the III-ring, underneath the peripheral stalk, a free volume is detected that is not accounted for by the detergent micelle (Figure 9.3 bottom, right). The free volume is a region with electron density in the 3D :map from electron cryo microscopy, in which no protein is found after placing the homology model. This volume is separated from the III-ring by a gap, which is much wider than the distance between the IIhring and subunit IV in the model. Therefore, the III14IV model was placed with subunit IV adjacent to the gap. The membrane parts of subunits I and II were modeled according to the N-terminal region of the E. coli subunit b ([20], Figure 9.3, bottom, centre) and were positioned into the free volume directly underneath the peripheral stalk (Figure 9.3, bottom right).

Was this article helpful?

0 0

Post a comment