Search

Translation initiation in eukaryotes requires the interplay of at least 10 initiation factors that interact at the different steps of this phase of gene expression. The interactions of initiation factors and related proteins are in general controlled by phosphorylation, which serves as a regulatory switch to turn protein translation on or off. The structures of initiation factors and a complete description of their post-translational modification (PTM) status are therefore required in order to fully understand these processes. In recent years, mass spectrometry has contributed considerably to provide this information and nowadays is proving to be indispensable when studying dynamic heterogeneous protein complexes such as the eukaryotic initiation factors. Herein, we highlight mass spectrometric approaches commonly applied to identify interacting subunits and their PTMs and the structural techniques that allow the architecture of protein complexes to be assessed. We present recent structural investigations of initiation factors and their interactions with other factors and with ribosomes and we assess the models generated. These models allow us to locate PTMs within initiation factor complexes and to highlight possible roles for phosphorylation sites in regulating interaction interfaces.

Homotypic death domain (DD)–DD interactions are important in the assembly of oligomeric signaling complexes such as the PIDDosome that acts as a platform for activation of caspase-2-dependent apoptotic signaling. The structure of the PIDDosome core complex exhibits an asymmetric three-layered arrangement containing five PIDD-DDs in one layer, five RAIDD-DDs in a second layer and an additional two RAIDD-DDs. We addressed complex formation between PIDD-DD and RAIDD-DD in solution using heteronuclear nuclear magnetic resonance (NMR) spectroscopy, nanoflow electrospray ionization mass spectrometry and size-exclusion chromatography with multi-angle light scattering. The DDs assemble into complexes displaying molecular masses in the range 130–158 kDa and RAIDD-DD:PIDD-DD stoichiometries of 5:5, 6:5 and 7:5. These data suggest that the crystal structure is representative of only the heaviest species in solution and that two RAIDD-DDs are loosely attached to the 5:5 core. Two-dimensional 1H,15N-NMR experiments exhibited signal loss upon complexation consistent with the formation of high-molecular-weight species. 13C-Methyl-transverse relaxation optimized spectroscopy measurements of the PIDDosome core exhibit signs of differential line broadening, cross-peak splitting and chemical shift heterogeneity that reflect the presence of non-equivalent sites at interfaces within an asymmetric complex. Experiments using a mutant RAIDD-DD that forms a monodisperse 5:5 complex with PIDD-DD show that the spectroscopic signature derives from the quasi- but non-exact equivalent environments of each DD. Since this characteristic was previously demonstrated for the complex between the DDs of CD95 and FADD, the NMR data for this system are consistent with the formation of a structure homologous to the PIDDosome core., Graphical Abstract, Highlights • The PIDDosome core particle that has been crystallized as a 7:5 complex displays heterogeneous stoichiometry in solution. • Methyl-transverse relaxation optimized spectroscopy NMR spectra for the complex suggest that individual PIDD-DDs and RAIDD-DDs experience non-equivalent environments in the PIDDosome core. • A mutant PIDDosome core particle that is monodisperse displays similar NMR features, suggesting that the complexity of the spectra is a reflection of the absence of formal symmetry consistent with the crystal structure. • The NMR characteristics are reminiscent of those reported for the complex formed between the DDs of CD95 and FADD, suggesting that this latter complex has similar architecture to the PIDDosome core.

DNA polymerase III, a decameric 420-kDa assembly, simultaneously replicates both strands of the chromosome in Escherichia coli. A subassembly of this holoenzyme, the seven-subunit clamp loader complex, is responsible for loading the sliding clamp (β2) onto DNA. Here, we use structural information derived from ion mobility mass spectrometry (IM-MS) to build three-dimensional models of one form of the full clamp loader complex, γ3δδ′ψχ (254 kDa). By probing the interaction between the clamp loader and a single-stranded DNA (ssDNA) binding protein (SSB4) and by identifying two distinct conformational states, with and without ssDNA, we assemble models of ψχ-SSB4 (108 kDa) and the clamp loader-SSB4 (340 kDa) consistent with IM data. A significant increase in measured collision cross-section (~ 10%) of the clamp loader-SSB4 complex upon DNA binding suggests large conformational rearrangements. This DNA bound conformation represents the active state and, along with the presence of ψχ, stabilises the clamp loader-SSB 4 complex. Overall, this study of a large heteromeric complex analysed by IM-MS, coupled with integrative modelling, highlights the potential of such an approach to reveal structural features of previously unknown complexes of high biological importance. © 2013 Elsevier Ltd.

The inherent heterogeneity of many protein assemblies complicates characterization of their structure and dynamics, as most biophysical techniques require homogeneous preparations of isolated components. For this reason, quantitative studies of the molecular chaperone αB-crystallin, which populates a range of interconverting oligomeric states, have been difficult, and the physicochemical basis for its polydispersity has remained unknown. Here, we perform mass spectrometry experiments to study αB-crystallin and extract detailed information as to its oligomeric distribution and exchange of subunits under a range of conditions. This allows a determination of the thermodynamic and kinetic parameters that govern the polydisperse ensemble and enables the construction of a simple energy profile for oligomerization. We find that the quaternary structure and dynamics of the protein can be explained using a simple model with just two oligomer-independent interactions (i.e., interactions that are energetically identical in all oligomers from 10mers to 40mers) between constituent monomers. As such, the distribution of oligomers is governed purely by the dynamics of individual monomers. This provides a new means for understanding the polydispersity of αB-crystallin and a framework for interrogating other heterogeneous protein assemblies.

The tetrameric tumor suppressor p53 plays a pivotal role in the control of the cell cycle and provides a paradigm for an emerging class of oligomeric, multidomain proteins with structured and intrinsically disordered regions. Many of its biophysical and functional properties have been extrapolated from truncated variants, yet the exact structural and functional role of certain segments of the protein is unclear. We found from NMR and X-ray crystallography that the DNA-binding domain (DBD) of human p53, usually defined as residues 94–292, extends beyond these domain boundaries. Trp91, in the hinge region between the disordered proline-rich N-terminal domain and the DBD, folds back onto the latter and has a cation–π interaction with Arg174. These additional interactions increase the melting temperature of the DBD by up to 2 °C and inhibit aggregation of the p53 tetramer. They also modulate the dissociation of the p53 tetramer. The absence of the Trp91/Arg174 packing presumably allows nonnative DBD–DBD interactions that both nucleate aggregation and stabilize the interface. These data have important implications for studies of multidomain proteins in general, highlighting the fact that weak ordered–disordered domain interactions can modulate the properties of proteins of complex structure., Graphical Abstract

T70N human lysozyme is the only known naturally occurring destabilised lysozyme variant that has not been detected in amyloid deposits in human patients. Its study and a comparison of its properties with those of the amyloidogenic variants of lysozyme is therefore important for understanding the determinants of amyloid disease. We report here the X-ray crystal structure and the solution dynamics of T70N lysozyme, as monitored by hydrogen/deuterium exchange and NMR relaxation experiments. The X-ray crystal structure shows that a substantial structural rearrangement results from the amino acid substitution, involving residues 45-51 and 68-75 in particular, and gives rise to a concomitant separation of these two loops of up to 6.5A. A marked decrease in the magnitudes of the generalised order parameter (S2) values of the amide nitrogen atom, for residues 70-74, shows that the T70N substitution increases the flexibility of the peptide backbone around the site of mutation. Hydrogen/deuterium exchange protection factors measured by NMR spectroscopy were calculated for the T70N variant and the wild-type protein. The protection factors for many of backbone amide groups in the beta-domain of the T70N variant are decreased relative to those in the wild-type protein, whereas those in the alpha-domain display wild-type-like values. In pulse-labelled hydrogen/deuterium exchange experiments monitored by mass spectrometry, transient but locally cooperative unfolding of the beta-domain of the T70N variant and the wild-type protein was observed, but at higher temperatures than for the amyloidogenic variants I56T and D67H. These findings reveal that such partial unfolding is an intrinsic property of the human lysozyme structure, and suggest that the readiness with which it occurs is a critical feature determining whether or not amyloid deposition occurs in vivo.

One of the 20 or so human amyloid diseases is associated with the deposition in vital organs of full-length mutational variants of the antibacterial protein lysozyme. Here, we report experimental data that permit a detailed comparison to be made of the behaviour of two of these amyloidogenic variants, I56T and D67H, under identical conditions. Hydrogen/deuterium exchange experiments monitored by NMR and mass spectrometry reveal that, despite their different locations and the different effects of the two mutations on the structure of the native state of lysozyme, both mutations cause a cooperative destabilisation of a remarkably similar segment of the structure, comprising in both cases the beta-domain and the adjacent C-helix. As a result, both variant proteins populate transiently a closely similar, partially unstructured intermediate in which the beta-domain and the adjacent C-helix are substantially and simultaneously unfolded, whereas the three remaining alpha-helices that form the core of the alpha-domain still have their native-like structure. We show, in addition, that the binding of a camel antibody fragment, cAb-HuL6, which was raised against wild-type lysozyme, restores to both variant proteins the stability and cooperativity characteristic of the wild-type protein; as a consequence, it inhibits the formation of amyloid fibrils by both variants. These results indicate that the reduction in global cooperativity, and the associated ability to populate transiently a specific, partly unfolded intermediate state under physiologically relevant conditions, is a common feature underlying the behaviour of these two pathogenic mutations. The formation of intermolecular interactions between lysozyme molecules that are in this partially unfolded state is therefore likely to be the fundamental trigger of the aggregation process that ultimately leads to the formation and deposition in tissue of amyloid fibrils.

The 22 members of the fibroblast growth factor (FGF) family have been implicated in cell proliferation, differentiation, survival, and migration. They are required for both development and maintenance of vertebrates, demonstrating an exquisite pattern of affinities for both protein and proteoglycan receptors. Recent crystal structures have suggested two models for the complex between FGFs, FGF receptors (FGFRs) and the proteoglycan heparan sulphate that mediates signalling, and have provided insight into how FGFs show differing affinities for the range of FGFRs. However, the physiological relevance of the two different models has not been made clear. Here, we demonstrate that the two complexes can be prepared from the same protein components, confirming that neither complex is the product of misfolded protein samples. Analyses of the complexes with mass spectrometry and analytical ultracentrifugation show that the species observed are consistent with the crystal structures formed using the two preparation protocols. This analysis supports the contention that both of the crystal structures reflect the state of the molecules in solution. Mass spectrometry of the complexes suggests that the stoichiometry of the complexes is 2 FGF1:2 FGFR2:1 heparin, regardless of the method used to prepare the complexes. These observations suggest that the two proposed complex architectures may both have relevance to the formation of an in vivo signalling complex, with a combination of the two interactions contributing to the formation of a larger focal complex.

The nuclear/hormone receptors are an extensive family of ligand-activated transcription factors that recognise DNA targets through a highly conserved, structurally autonomous DNA-binding domain. The compact structure of the DNA-binding domain is supported by two zinc ions, each of which is co-ordinated by the tetrahedral arrangement of thiol groups from four cysteine residues. Metal binding is expected to be linked with deprotonation of the co-ordinating thiol groups and folding of the polypeptide. Using a variety of biophysical approaches, we characterise these linked equilibria for the isolated DNA-binding domains (DBD) of the receptors for estrogen and glucocorticoid. Mass spectrometry and equilibrium denaturation indicate that, near neutral pH, approximately four of the eight co-ordinating thiol groups release protons with zinc uptake, in agreement with the expected p K a change for the –SH group in the presence of the metal. Mass spectrometry reveals that the protein charge distribution changes with the uptake of zinc and that metal binding is co-operative. The co-operativity is consistent with observations from equilibrium denaturation, which indicate that the folding event is a two-state process. A crucial residue that stabilises the equilibrium structure of the DBD fold itself is a cysteine residue situated in the hydrophobic core of all known nuclear hormone receptors (but not involved in metal binding): it appears to be conserved absolutely for its unique combination of size and hydrophobicity. Stabilisation of the DBDs could be achieved by truncating the flexible, basic termini, suggesting that like-charge clusters may have deleterious effects on protein folds. While the metal-free apo protein and the chemically denatured state have little defined secondary structure, these states were expanded only partially in comparison with the native structure, according to data from small-angle X-ray scattering. The comparatively compact shapes of the denatured and apo forms may explain, in part, the marginal stability of the native fold.

Mass spectrometry (MS) was used to characterise the binding of the 58 kDa protein OppA to 11 peptides with diverse properties. Peptides with two, three and five amino acid residues were added to OppA, and the mass spectra showed that the highest-affinity complexes are formed between OppA and tripeptide ligands. Lower-affinity complexes were observed for OppA and dipeptide ligands, and no complex formation was detected with pentapeptides or a tripeptide in which the N-terminal amino group was acetylated. Tripeptides containing a single d amino acid residue were found not to bind to native OppA. Evidence from the peak width and the, charge in the spectra of the complexes suggests that the bound peptides are encapsulated by the protein in a solvent-filled cavity in the gas phase of the mass spectrometer. Analysis of the proportions of peptide-bound and free proteins under low-energy MS conditions shows a good correlation with solution-phase K(d) measurements where available. Increasing the internal energy of the gas-phase complex led to dissociation of the complex. The ease of dissociation is interpreted in terms of the intrinsic stability of the complex in the absence of the stabilising effects of bulk solvent. The results from this study demonstrate insensitivity to the hydrophobic and ionic properties, of the side-chains of the peptides, in contrast to the investigation of other protein ligand systems by MS. Moreover, these findings are in accord with the physiological role of this protein in allowing into the cell di- and tripeptides containing naturally occurring amino acids, regardless of their sequence, while barring access to potentially harmful peptide mimics.

The refolding of bovine alpha-lactalbumin (BLA) from its chemically denatured state in 6 M GuHCl has been investigated by a variety of complementary biophysical approaches. CD experiments indicate that the species formed in the early stages of refolding of the apo-protein have at least 85 % of the alpha-helical content of the native state, and kinetic NMR experiments show that they possess near-native compactness. Hydrogen exchange measurements using mass spectrometry and NMR indicate that persistent structure in these transient species is located predominantly in the alpha-domain of the native protein and is similar to that present in the partially folded A-state formed by the protein at low pH. The extent of the exchange protection is, however, small, and there is no evidence for the existence of well-defined discrete kinetic intermediates of the type populated in the refolding of the structurally homologous c-type lysozymes. Rather, both mass spectrometric and NMR data indicate that the rate-determining transition from the compact partially structured (molten globule) species to the native state is highly cooperative. The data show that folding in the presence of Ca2+ is similar to that in its absence, although the rate is increased by more than two orders of magnitude. Sequential mixing experiments monitored by fluorescence spectroscopy indicate that this slower folding is not the result of the accumulation of kinetically trapped species. Rather, the data are consistent with a model in which binding of Ca2+ stabilizes native-like contacts in the partially folded species and reduces the barriers for the conversion of the protein to its native state. Taken together the results indicate that folding of BLA, in the presence of its four disulphide bonds, corresponds to one of the limiting cases of protein folding in which rapid collapse to a globule with a native-like fold is followed by a search for native-like side-chain contacts that enable efficient conversion to the close packed native structure.

Reduced denatured lysozyme has been oxidised and refolded at pH values close to neutral in an efficient way by dilution from buffers containing 8.0 M urea, and refolding intermediates were separated by reverse-phase HPLC at pH 2. By using peptic digestion in combination with high-resolution Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometry (MS) and tandem MS/MS the dominant intermediate was identified to be des-[76–94]. This species has three of the four native disulphide bonds, but lacks the Cys76-Cys94 disulphide bond which connects the two folding domains in the native protein. Characterisation of des-[76–94] by 2D 1H NMR shows that it has a highly native-like structure. This provides an explanation for the accumulation of this species during refolding as direct oxidation to the fully native protein will be restricted by the burial of Cys94 in the protein interior.

That membrane protein complexes could survive in the gas phase had always seemed impossible. The lack of chargeable residues, high hydrophobicity, and poor solubility and the vast excess of detergent contributed to the view that it would not be possible to obtain mass spectra of intact membrane complexes. With the recent success in recording mass spectra of these complexes, first from recombinant sources and later from the cellular environment, many surprising properties of these gas phase membrane complexes have been revealed. The first of these was that the interactions between membrane and soluble subunits could survive in vacuum, without detergent molecules adhering to the complex. The second unexpected feature was that their hydrophobicity and, consequently, lower charge state did not preclude ionization. The final surprising finding was that these gas phase membrane complexes carry with them lipids, bound specifically in subunit interfaces. This provides us with an opportunity to distinguish annular lipids that surround the membrane complexes, from structural lipids that have a role in maintaining structure and subunit interactions. In this perspective, we track these developments and suggest explanations for the various discoveries made during this research.

The ribosomal stalk complex in Escherichia coli consists of L10 and four copies of L7/L12, and is largely responsible for binding and recruiting translation factors. Structural characterisation of this stalk complex is difficult, primarily due to its dynamics. Here, we apply mass spectrometry to follow post-translational modifications and their effect on structural changes of the stalk proteins on intact ribosomes. Our results show that increased acetylation of L12 occurs during the stationary phase on ribosomes harvested from cells grown under optimal conditions. For cells grown in minimal medium, L12 acetylation and processing is altered, resulting in deficient removal of N-terminal methionine in approximately 50% of the L12 population, while processed L12 is almost 100% acetylated. Our results show also that N-acetylation of L12 correlates with an increased stability of the stalk complex in the gas phase. To investigate further the basis of this increased stability, we applied a solution phase hydrogen deuterium exchange protocol to compare the rate of deuterium incorporation in the proteins L9, L10, L11 and L12 as well as the acetylated form of L12 (L7), in situ on the ribosome. Results show that deuterium incorporation is consistently slower for L7 relative to L12 and for L10 when L7 is predominant. Our results imply a tightening of the interaction between L7 and L10 relative to that between L12 and L10. Since acetylation is predominant when cells are grown in minimal medium, we propose that these modifications form part of the cell's strategy to increase stability of the stalk complex under conditions of stress. More generally, our results demonstrate that it is possible to discern the influence of a 42 Da post-translational modification by mass spectrometry and to record subtle changes in hydrogen/deuterium exchange within the context of an intact 2.5 MDa particle.

The molecular chaperone Hsp90 is essential for the correct folding, maturation and activation of a diverse array of client proteins, including several key constituents of oncogenic processes. Hsp90 has become a focus of cancer research, since it represents a target for direct prophylaxis against multistep malignancy. Hydrogen-exchange mass spectrometry was used to study the structural and conformational changes undergone by full-length human Hsp90beta in solution upon binding of the kinase-specific co-chaperone Cdc37 and two Hsp90 ATPase inhibitors: Radicicol and the first-generation anticancer drug DMAG. Changes in hydrogen exchange pattern in the complexes in regions of Hsp90 remote to the ligand-binding site were observed indicating long-range effects. In particular, the interface between the N-terminal domain and middle domains exhibited significant differences between the apo and complexed forms. For the inhibitors, differences in the interface between the middle domain and the C-terminal domain were also observed. These data provide important insight into the structure of the biologically active form of the protein.

Protein oligomerisation is a prerequisite for the toxicity of a number of bacterial toxins. Examples include the pore-forming cytotoxin streptolysin O, which oligomerises to form large pores in the membrane and the protective antigen of anthrax toxin, where a heptameric complex is essential for the delivery of lethal factor and edema factor to the cell cytosol. Binding of the clostridial neurotoxins to receptors on neuronal cells is well characterised, but little is known regarding the quaternary structure of these toxins and the role of oligomerisation in the intoxication process. We have investigated the oligomerisation of the receptor binding domain (H(C)) of tetanus toxin, which retains the binding and trafficking properties of the full-length toxin. Electrophoresis, size exclusion chromatography and mass spectrometry were used to demonstrate that H(C) undergoes concentration-dependent oligomerisation in solution. Reducing agents were found to affect H(C) oligomerisation and, using mutagenesis, Cys869 was shown to be essential for this process. Furthermore, the oligomeric state and quaternary structure of H(C) in solution was assessed using synchrotron small-angle X-ray scattering. Ab initio shape analysis and rigid body modelling coupled with mutagenesis data allowed the construction of an unequivocal model of dimeric H(C) in solution. We propose a possible mechanism for H(C) oligomerisation and discuss how this may relate to toxicity.

Amyloid fibrils are typically rigid, unbranched structures with diameters of approximately 10 nm and lengths up to several micrometres, and are associated with more than 20 diseases including Alzheimer's disease and type II diabetes. Insulin is a small, predominantly alpha-helical protein consisting of 51 residues in two disulfide-linked polypeptide chains that readily assembles into amyloid fibrils under conditions of low pH and elevated temperature. We demonstrate here that both the A-chain and the B-chain of insulin are capable of forming amyloid fibrils in isolation under similar conditions, with fibrillar morphologies that differ from those composed of intact insulin. Both the A-chain and B-chain fibrils were found to be able to cross-seed the fibrillization of the parent protein, although these reactions were substantially less efficient than self-seeding with fibrils composed of full-length insulin. In both cases, the cross-seeded fibrils were morphologically distinct from the seeding material, but shared common characteristics with typical insulin fibrils, including a very similar helical repeat. The broader distribution of heights of the cross-seeded fibrils compared to typical insulin fibrils, however, indicates that their underlying protofilament hierarchy may be subtly different. In addition, and remarkably in view of this seeding behavior, the soluble forms of the A-chain and B-chain peptides were found to be capable of inhibiting insulin fibril formation. Studies using mass spectrometry suggest that this behavior might be attributable to complex formation between insulin and the A-chain and B-chain peptides. The finding that the same chemical form of a polypeptide chain in different physical states can either stimulate or inhibit the conversion of a protein into amyloid fibrils sheds new light on the mechanisms underlying fibril formation, fibril strain propagation and amyloid disease initiation and progression.

The action of the molecular chaperone Hsp90 is essential for the activation and assembly of an increasing number of client proteins. This function of Hsp90 has been proposed to be governed by conformational changes driven by ATP binding and hydrolysis. Association of co-chaperones and client proteins regulate the ATPase activity of Hsp90. Here, we have examined the inhibition of the ATPase activity of human Hsp90beta by one such co-chaperone, human p23. We demonstrate that human p23 interacts with Hsp90 in both the absence and presence of nucleotide with a higher affinity in the presence of the ATP analogue AMP-PNP. This is consistent with an analysis of the effect of p23 on the steady-state kinetics that revealed a mixed mechanism of inhibition. Mass spectrometry of the intact Hsp90.p23 complex determined the stoichiometry of binding to be one p23 to each subunit of the Hsp90 dimer. p23 was also shown to interact with a monomeric, truncated fragment of Hsp90, lacking the C-terminal homodimerisation domain, indicating dimerisation of Hsp90 is not a prerequisite for association with p23. Complex formation between Hsp90 and p23 increased the apparent affinity of Hsp90 for AMP-PNP and completely inhibited the ATPase activity. We propose a model where the role of p23 is to lock individual subunits of Hsp90 in an ATP-dependent conformational state that has a high affinity for client proteins.

The spliceosomal U1C protein is critical to the initiation and regulation of precursor messenger RNA (pre-mRNA) splicing, as part of the U1 small nuclear ribonucleoprotein particle (snRNP). We have produced full-length and 61 residue constructs of human U1C in soluble form in Escherichia coli. Atomic absorption spectroscopy and mass spectrometry show that both constructs contain one Zn atom and are monomeric. Gelmobility-shift assays showed that one molecule of recombinant U1C, either full-length or 61 residue construct, can be incorporated into the U1 snRNP core domain in the presence of U1 70k. This result is in perfect agreement with the previous experiment with U1C isolated from the HeLa U1 snRNP showing that the recombinant U1C is functionally active. We have determined the solution structure of the N-terminal 61 residue construct of U1C by NMR. A Cys(2)His(2)-type zinc finger, distinct from the TFIIIA-type, is extended at its C terminus by two additional helices. The two Zn-coordinating histidine residues are separated by a five residue loop. The conserved basic residues in the first two helices and the intervening loop may be involved in RNA binding. The opposite beta-sheet face with two surface-exposed Tyr residues may be involved in protein contacts. Both the full-length and 61 residue constructs of human U1C fail to bind RNA containing the 5' splice site sequence, in contrast to what has been reported for the Saccharomyces cerevisiae orthologue.

The heat shock protein Hsp90 plays a key, but poorly understood role in the folding, assembly and activation of a large number of signal transduction molecules, in particular kinases and steroid hormone receptors. In carrying out these functions Hsp90 hydrolyses ATP as it cycles between ADP- and ATP-bound forms, and this ATPase activity is regulated by the transient association with a variety of co-chaperones. Cdc37 is one such co-chaperone protein that also has a role in client protein recognition, in that it is required for Hsp90-dependent folding and activation of a particular group of protein kinases. These include the cyclin-dependent kinases (Cdk) 4/6 and Cdk9, Raf-1, Akt and many others. Here, the biochemical details of the interaction of human Hsp90β and Cdc37 have been characterised. Small angle X-ray scattering (SAXS) was then used to study the solution structure of Hsp90 and its complexes with Cdc37. The results suggest a model for the interaction of Cdc37 with Hsp90, whereby a Cdc37 dimer binds the two N-terminal domain/linker regions in an Hsp90 dimer, fixing them in a single conformation that is presumably suitable for client protein recognition.

The hydrogen-exchange behavior of the low-pH molten globule of human alpha-lactalbumin, containing all four disulfides, has been examined and compared with that of a single disulfide variant, [28-111] alpha-lactalbumin, and of a series of proline variants of [28-111] alpha-lactalbumin. The small differences in hydrogen-exchange protection exhibited by these partially folded species were compared by mixing two or more proteins and monitoring their exchange simultaneously using mass spectrometry. The effect of single proline mutations within each alpha-domain helix on hydrogen-exchange protection has been investigated using six proline variants of [28-111] alpha-lactalbumin, L11P, L12P, M30P, I95P, K108P and Q117P. The results show that proline mutations in the A, B, C and D alpha-helices lead to a loss of hydrogen-exchange protection for residues in the local helix without perturbing hydrogen-exchange protection in other regions of the protein. Thus, local unfolding of the A, B, C and D helices does not significantly alter the packing and solvent accessibility of other regions of the molten globule. By contrast, introduction of a proline residue in the C-terminal 3(10) helix produces a larger and more widespread loss of hydrogen-exchange protection, demonstrating that longer-range perturbations of the molten globule have occurred. Thus, residues in this C-terminal region must be involved in contacts that are critical for the stabilisation of the compact molten globule structure.

Wild-type hen lysozyme has been converted from its soluble native state into highly organized amyloid fibrils. In order to achieve this conversion, conditions were chosen to promote partial unfolding of the native globular fold and included heating of low-pH solutions and addition of organic solvents. Two peptides derived from the beta-sheet region of hen lysozyme were also found to form fibrils very readily. The properties and morphologies of the amyloid fibrils formed by incubation either of the protein or the peptides are similar to those produced from the group of proteins associated with clinical amyloidoses. Fibril formation by hen lysozyme was substantially accelerated when aliquots of solutions in which fibrils of either one of the peptides or the full-length protein had previously formed were added to fresh solutions of the protein, revealing the importance of seeding in the kinetics of fibril formation. These findings support the proposition that the beta-domain is of particular significance in the formation of fibrils from the full-length protein and suggest similarities between the species giving rise to fibril formation and the intermediates formed during protein folding.

A variety of techniques, including quenched-flow hydrogen exchange labelling monitored by electrospray ionization mass spectrometry, and stopped-flow absorbance, fluorescence and circular dichroism spectroscopy, has been used to investigate the refolding kinetics of hen lysozyme over a temperature range from 2 degrees C to 50 degrees C. Simple Arrhenius behaviour is not observed, and although the overall rate of folding increases from 2 to 40 degrees C, it decreases above 40 degrees C. In addition, the transient intermediate on the major folding pathway at 20 degrees C, in which the alpha-domain is persistently structured in the absence of a stable beta-domain, is thermally unfolded in a sigmoidal transition (T(m) approximately 40 degrees C) indicative of a cooperatively folded state. At all temperatures, however, there is evidence for fast ( approximately 25 %) and slow ( approximately 75 %) populations of refolding molecules. By using transition state theory, the kinetic data from various experiments were jointly fitted to a sequential three-state model for the slow folding pathway. Together with previous findings, these results indicate that the alpha-domain intermediate is a productive species on the folding route between the denatured and native states, and which accumulates as a consequence of its intrinsic stability. Our analysis suggests that the temperature dependence of the rate constant for lysozyme folding depends on both the total change in the heat capacity between the ground and transition states (the dominant factor at low temperatures) and the heat-induced destabilization of the alpha-domain intermediate (the dominant factor at high temperatures). Destabilization of such kinetically competent intermediate species is likely to be a determining factor in the non-Arrhenius temperature dependence of the folding rate of those proteins for which one or more intermediates are populated.

Wild-type and variant transthyretins form amyloid fibrils in two different diseases. The biologically active form of transthyretin is a tetramer but there is evidence that a monomeric species is the amyloidogenic intermediate. Using mass spectrometry we have developed an approach to monitor the proportions of monomer and tetramer in wild-type and variant transthyretins, and found a strong correlation between the instability of the tetramer in the gas phase and the amyloidogenicity of the protein variant. The presence of water molecules in the central channel has been found to be critical for maintaining intact the complex in the gas phase, with additional stability observed in the presence of excess thyroxine. The solution structure of monomeric transthyretin under fibril-forming conditions was studied using hydrogen exchange monitored by mass spectrometry. The results show that Val30Met transthyretin, the commonest amyloidogenic variant, exhibits loss of hydrogen exchange protection substantially more rapidly than the wild-type protein, suggesting partial unfolding of the β-sheet structure. These results provide new insights into the correlation between tetramer stability and amyloidogenicity as well as supporting a possible route to fibril formation via transient unfolding of the transthyretin monomer.