Analysis of the interaction between the eukaryotic chaperonin CCT and its substrates actin and tubulin.

Two mechanisms have thus far been characterized for the assistance by chaperonins of the folding of other proteins. The first and best described is that of the prokaryotic chaperonin GroEL, which interacts with a large spectrum of proteins. GroEL uses a nonspecific mechanism by which any conformation of practically any unfolded polypeptide interacts with it through exposed, hydrophobic residues. ATP binding liberates the substrate in the GroEL cavity where it is given a chance to fold. A second mechanism has been described for the eukaryotic chaperonin CCT, which interacts mainly with the cytoskeletal proteins actin and tubulin. Cryoelectron microscopy and biochemical studies have revealed that both of these proteins interact with CCT in quasi-native, defined conformations. Here we have performed a detailed study of the docking of the actin and tubulin molecules extracted from their corresponding CCT:substrate complexes obtained from cryoelectron microscopy and image processing to localize certain regions in actin and tubulin that are involved in the interaction with CCT. These regions of actin and tubulin, which are not present in their prokaryotic counterparts FtsA and FtsZ, are involved in the polymerization of the two cytoskeletal proteins. These findings suggest coevolution of CCT with actin and tubulin in order to counteract the folding problems associated with the generation in these two cytoskeletal protein families of new domains involved in their polymerization.

The 'sequential allosteric ring' mechanism in the eukaryotic chaperonin-assisted folding of actin and tubulin.

Folding to completion of actin and tubulin in the eukaryotic cytosol requires their interaction with cytosolic chaperonin CCT [chaperonin containing tailless complex polypeptide 1 (TCP-1)]. Three-dimensional reconstructions of nucleotide-free CCT complexed to either actin or tubulin show that CCT stabilizes both cytoskeletal proteins in open and quasi-folded conformations mediated through interactions that are both subunit specific and geometry dependent. Here we find that upon ATP binding, mimicked by the non-hydrolysable analog AMP-PNP (5'-adenylyl-imido-diphosphate), to both CCT-alpha-actin and CCT- beta-tubulin complexes, the chaperonin component undergoes concerted movements of the apical domains, resulting in the cavity being closed off by the helical protrusions of the eight apical domains. However, in contrast to the GroE system, generation of this closed state does not induce the release of the substrate into the chaperonin cavity, and both cytoskeletal proteins remain bound to the chaperonin apical domains. Docking of the AMP-PNP-CCT-bound conformations of alpha-actin and beta-tubulin to their respective native atomic structures suggests that both proteins have progressed towards their native states.

Defining the eukaryotic cytosolic chaperonin-binding sites in human tubulins.

The actins and tubulins are the obligate substrates in vivo of the chaperonin-containing TCP-1 (CCT). The precise elements of recognition between the chaperonin and its substrates remain largely unknown. We have used a solid phase peptide binding assay to screen the human alpha, beta and gamma-tubulin sequences for CCT recognition. Multiple regions seem to be implicated in interactions between tubulins and CCT. These potential CCT-binding sites are highly dispersed throughout the primary sequences of the human tubulins. In addition, using site-directed mutagenesis we assessed the contribution of the selected residues in the C-terminal domain of beta-tubulin to CCT binding. Various hot spots have been identified even though, in each case, their replacement by alanine does not reduce dramatically the total affinity of beta-tubulin for CCT. The CCT-binding information in the tubulins is probably confined to multiple specific regions each having weak or moderate affinity for CCT apical domains. The main binding region seems to be located between residues 263 and 384, but there are no single amino acid residues in this region, which make large contributions to the binding energy, although we have detected a minor contribution by F377. These biochemical results are understandable in the context of our recent structural analysis of CCT-tubulin complexes by cryo-electron microscopy and image reconstruction, which shows that, in one stage of an in vitro binding reaction between apo-CCT and tubulin diluted from guanidinium chloride, ten major, stable contacts between tubulin and CCT are involved. Therefore, specificity is achieved through the co-operation of many specific, albeit weak, interactions.

Eukaryotic chaperonin CCT stabilizes actin and tubulin folding intermediates in open quasi-native conformations.

Three-dimensional reconstruction from cryoelectron micrographs of the eukaryotic cytosolic chaperonin CCT complexed to tubulin shows that CCT interacts with tubulin (both the alpha and beta isoforms) using five specific CCT subunits. The CCT-tubulin interaction has a different geometry to the CCT-actin interaction, and a mixture of shared and unique CCT subunits is used in binding the two substrates. Docking of the atomic structures of both actin and tubulin to their CCT-bound conformation suggests a common mode of chaperonin-substrate interaction. CCT stabilizes quasi-native structures in both proteins that are open through their domain-connecting hinge regions, suggesting a novel mechanism and function of CCT in assisted protein folding.

Is the continuity of the domains required for the correct folding of a two-domain protein?

The role of domains in protein folding has been widely studied and discussed. Nevertheless, it is not clear whether the continuity of the domains in a protein is an essential requirement in determining the folding pathway. Previous studies have shown that the isolated structural domains of the two-domain monomeric enzyme, yeast phosphoglycerate kinase (yPGK), are able to fold independently into a quasinative structure, but they neither reassociate nor generate a functional enzyme [Minard, P., Hall, L., Betton, J. M., Missiakas, D., & Yon, J. M. (1989) Protein Eng. 3, 55-60; Fairbrother, W. J., Bowen, D., Hall, L., Williams, R. J. P. (1989) Eur. J. Biochem. 184, 617-625; Missiakas, D., Betton, J. M., Minard, P., & Yon, J. M. (1990) Biochemistry 29, 8683-8689]. In the present work, two circularly permuted variants of the yPGK gene were constructed. The natural adjacent chain termini were directly connected and the new extremities were created within the N-domain (at residues 71 and 72) or the C-domain (at residues 291 and 292), respectively. These two proteins were overexpressed and purified. They exhibit 14% and 23% of the wild-type enzyme activity, respectively. The two mutants fold in a compact conformation with slight changes in the secondary and tertiary structure probably related to the presence of a heterogeneous population of molecules. The unfolding studies reveal a large decrease in stability. From the present data it appears that, although the circular permutations induce some perturbations in the structure and stability of the protein, the continuity of the domains is not required for the protein to reach a native-like and functional structure.

Role of the C-terminal helix in the folding and stability of yeast phosphoglycerate kinase.

In order to determine the role of the C-terminal helix in the folding and stability of yeast phosphoglycerate kinase, a mutant deleted of the 12 C-terminal residues (PGK delta 404-415) was constructed. This mutant folds in a conformation very similar to that of the wild-type protein, but exhibits a very low activity (0.1% of that of the wild-type enzyme). The main structural effect of the deletion of the C-terminal helix is an increase in flexibility of the whole protein and a decrease in stability by about 5 kcal/mol. The structural properties of the truncated protein are very similar, at least qualitatively, to those in the isolated domains. The accessibility of the thiol group of Cys 97 is identical to that in the isolated N-domain. The large solvent effect on the tryptophan fluorescence in the native protein at very low concentration of denaturant reveals an increase of flexibility of the C-domain, similar to that observed on the isolated C-domain. NMR measurements show that the pH dependence of His C2H and C4H chemical shifts in the truncated protein perfectly matches those of the isolated domains. The addition of the missing peptide provokes a 40-fold increase in enzyme activity at saturation. A dissociation constant of 80 microM was determined. This peptide, which displays a random structure in solution, folds in a helical structure in the region 405-410 as assessed by TRNOESY. All these results show that the C-terminal part of yeast phosphoglycerate kinase is not necessary for most of the initial folding steps but acts to lock the C-domain on the N-domain, thus ensuring the expression of full enzyme activity. Without this sequence, the protein has the sum of the properties of the two isolated domains.

Transferred nuclear Overhauser effect study of the C-terminal helix of yeast phosphoglycerate kinase: NMR solution structure of the C-terminal bound peptide.

Two-dimensional 1H nuclear magnetic resonance spectroscopy is used to determine the structure of the C-terminal complementary peptide (404-415) bound to a mutant phosphoglycerate kinase (1-403). Conformational changes in the peptide induced by the formation of the peptide-protein complex are followed by transferred nuclear Overhauser effect spectroscopy. Measurement of transferred NOEs and molecular modeling reveal an alpha-helix fold in the 405-409 region. This fold is in good agreement with the corresponding helix XIV of the crystallographic structure of wild-type PGK (Watson et al., 1982). The role of the alpha-helix from the C-terminal peptide in the recovery of catalytic activity in the mutant PGK is discussed.