Chaperones of F1-ATPase

Mitochondrial F₁-ATPase contains a hexamer of alternating α and β subunits. The assembly of this structure requires two specialized chaperones, Atp11p and Atp12p, that bind transiently to β and α. In the absence of Atp11p and Atp12p, the hexamer is not formed, and α and β precipitate as large insoluble aggregates. An early model for the mechanism of chaperone-mediated F₁ assembly (Wang, Z. G., Sheluho, D., Gatti, D. L., and Ackerman, S. H. (2000) EMBO J. 19, 1486–1493) hypothesized that the chaperones themselves look very much like the α and β subunits, and proposed an exchange of Atp11p for α and of Atp12p for β; the driving force for the exchange was expected to be a higher affinity of α and β for each other than for the respective chaperone partners. One important feature of this model was the prediction that as long as Atp11p is bound to β and Atp12p is bound to α, the two F₁ subunits cannot interact at either the catalytic site or the noncatalytic site interface. Here we present the structures of Atp11p from Candida glabrata and Atp12p from Paracoccus denitrificans, and we show that some features of the Wang model are correct, namely that binding of the chaperones to α and β prevents further interactions between these F₁ subunits. However, Atp11p and Atp12p do not resemble α or β, and it is instead the F₁ γ subunit that initiates the release of the chaperones from α and β and their further assembly into the mature complex.Mitochondrial F₁-ATPase consists of three α and three β subunits occupying alternate positions in a hexamer that surrounds a rod-like element containing one each of γ, δ, and ϵ subunits (1–3). Three nucleotide-binding catalytic sites (CS)⁴ and three noncatalytic sites (NCS) alternate at the six α/β interfaces. Early work with respiratory-deficient strains of Saccharomyces cerevisiae (4) revealed that two additional mitochondrial proteins, Atp11p and Atp12p, which are not integral subunits of the enzyme, are nonetheless necessary for the assembly of F₁-ATPase. Besides their failure to assemble F₁, a particularly interesting feature of atp11 and atp12 mutants is that they accumulate α and β subunits as high molecular weight aggregates (4) that can be recognized as densely stained inclusion bodies in the mitochondrial matrix (5). Subsequent studies in yeast have shown that Atp12p binds to F₁ α (6) and that Atp11p binds to β (7); these interactions include binding determinants in the nucleotide binding domains (NBD) of the two F₁ subunits. On this basis, it is now recognized that Atp11p and Atp12p are members of two new families of molecular chaperones, pfam06644 and pfam07542 (8), which are required for the assembly of mitochondrial ATP synthase in all eukaryotes. In fact, the first nuclear genetic lesion associated to a defect of mitochondrial ATP synthase in humans was identified in the locus ATPAF2 for Atp12p and was responsible for the death of a 14-month-old infant (9). Atp12p is also present in the α subdivision of Proteobacteria, consistent with the proposed origin of mitochondria from this ancestral line (10).The nature of the interactions between the F₁ subunits and Atp11p and Atp12p has remained elusive because of the lack of structural information for these chaperones. As α and β aggregate in the absence of Atp11p and Atp12p, it is usually assumed that the F₁ subunits are themselves poorly soluble, and that the two chaperones maintain them in a dispersed state until they are incorporated in the mature enzyme. Based on the analysis of the distribution of hydrophilic and hydrophobic areas on the surface of the α and β subunits of F₁, and on the interaction energies between these subunits at the interfaces that provide the CS and NCS sites, Wang et al. (6) have proposed a model of F₁ assembly in which Atp11p binds at the region of the β subunit that contributes to the CS site, and Atp12p binds at the region of the α subunit that contributes to the NCS site. One consequence of this particular binding of Atp11p and Atp12p to the F₁ subunits is that as long as Atp11p is bound to β and Atp12p is bound to α, the two F₁ subunits cannot interact at either the CS or the NCS interface. Since no other modulators of chaperone release are known, the Wang model requires an exchange of Atp11p for α and of Atp12p for β. Implied in this model is that the chaperones must themselves look very much like the α and β subunits, and that the driving force for the exchange must simply be a higher affinity of α and β for each other than for the respective chaperone partners. Here we present the structures of Atp11p from Candida glabrata and Atp12p from Paracoccus denitrificans, and we show that some features of the Wang model are correct, namely that binding of the chaperones to α and β prevents further interactions between these F₁ subunits. However, Atp11p and Atp12p do not resemble α or β, and it is instead the F₁ γ subunit that initiates the release of the chaperones from α and β and their further assembly into mature complex.