Search

Using protein from the hyperthermophile Thermotoga maritima, we have determined the solution structure of CheW, an essential component in the formation of the bacterial chemotaxis signaling complex. The overall fold is similar to the regulatory domain of the chemotaxis kinase CheA. In addition, interactions of CheW with CheA were monitored by nuclear magnetic resonance (NMR) techniques. The chemical shift perturbation data show the probable contacts that CheW makes with CheA. In combination with previous genetic data, the structure also suggests a possible binding site for the chemotaxis receptor. These results provide a structural basis for a model in which CheW acts as a molecular bridge between CheA and the cytoplasmic tails of the receptor.

Protein structure is inherently dynamic, with function often predicated on excursions from low to higher energy conformations. For example, X-ray studies of a cavity mutant of T4 lysozyme, L99A, show that the cavity is sterically inaccessible to ligand, yet the protein is able to bind substituted benzenes rapidly. We have used novel relaxation dispersion NMR techniques to kinetically and thermodynamically characterize a transition between a highly populated (97%, 25 degrees C) ground state conformation and an excited state that is 2.0 kcal mol(-1) higher in free energy. A temperature-dependent study of the rates of interconversion between ground and excited states allows the separation of the free energy change into enthalpic (Delta H = 7.1 kcal mol(-1)) and entropic (T Delta S = 5.1 kcal mol(-1), 25 degrees C) components. The residues involved cluster about the cavity, providing evidence that the excited state facilitates ligand entry.

We have used native state exchange to examine the energy landscape of the well-characterized protein T4 lysozyme. Although the protein exhibits two-state behavior by traditional probes, the energy landscape determined here is much more complex. The average stability of the C-terminal subdomain is significantly higher than that for the N-terminus suggesting at least two regions of unfolding. At a more detailed level, there appears to be a broad continuum of stabilities throughout each region. The overall subdomain hierarchy of energies does not mirror data on the folding pathway for this protein, challenging the relationship between energy landscapes and folding trajectories.

The interaction between the flagellum specific sigma factor, sigma 28, and its inhibitor, FlgM, was examined using multidimensional heteronuclear NMR. Here we observe that free FlgM is mostly unfolded, but about 50% of the residues become structured when bound to sigma 28. Our analysis suggests that the sigma 28 binding domain of FlgM is contained within the last 57 amino acids of the protein while the first 40 amino acids are unstructured in both the free and bound states. Genetic analysis of flgM mutants that fail to inhibit sigma 28 activity reveal amino acid changes that are also isolated to the C-terminal 57 residues of FlgM. We postulate that the lack of structure in free and bound FlgM is important to its role as an exported protein.

Human haemoglobin has long been a paradigm for cooperative ligand binding and allostery. Through analysis of the crystal structures of deoxyhaemoglobin and liganded haemoglobin, Perutz proposed a model for cooperativity in which the bond between the proximal histidine and the protein couples haem rearrangements to protein structure rearrangements. Here we test this model by deleting the bonds between the histidine imidazole side chain and the polypeptide. This detachment method allows us to determine directly the contribution of proximal histidine coupling to cooperativity of distal ligand binding. Proximal detachment significantly increases ligand affinity, reduces cooperativity, and prevents quaternary structure switching, in accord with the Perutz model. Residual cooperativity indicates that additional haem communication pathways exist that do not involve the proximal histidine coupling mechanism.

We have investigated the magnitude and timescale of fluctuations within the core of a protein using the exchange kinetics of indole and benzene binding to engineered hydrophobic cavities in T4 lysozyme. The crystal structures of variant-benzene complexes suggest that relatively large scale fluctuations (1-2 angstrom) of backbone atoms are required for entry of these ligands into the core. Nonetheless, these ligands enter the cavities rapidly, with bimolecular rate constants of approximately 10(6)-10(7) M(-1) s(-1) and a low activation barrier, 2-5 kcal mol(-1). These results suggest that protein cores undergo substantial fluctuations on the millisecond to microsecond timescale and that entry of small molecules into protein interiors is not strongly limited by steric occlusion.

The tertiary fold of the epsilon subunit of the Escherichia coli F1F0 ATPsynthase (ECF1F0) has been determined by two- and three-dimensional heteronuclear (13C, 15N) NMR spectroscopy. The epsilon subunit exhibits a distinct two domain structure, with the N-terminal 84 residues of the protein forming a 10-stranded beta-structure, and with the C-terminal 48 amino acids arranged as two alpha-helices running antiparallel to one another (two helix hairpin). The beta-domain folds as a beta-sandwich with a hydrophobic interior between the two layers of the sandwich. The C-terminal two-helix hairpin folds back to the N-terminal domain and interacts with one side of the beta-domain. The arrangement of the epsilon subunit in the intact F1F0 ATP synthase involves interaction of the two helix hairpin with the F1 part, and binding of the open side of the beta-sandwich to the c subunits of the membrane-embedded F0 part.

Phosphotransfer between the autophosphorylating histidine kinase CheA and the response regulator CheY represents a crucial step in the bacterial chemotaxis signal transduction pathway. The 15N-1H correlation spectrum of CheY complexed with an amino-terminal fragment of CheA exhibits specific localized differences in chemical shifts when compared to the spectrum of uncomplexed CheY. When mapped onto the three-dimensional structure of CheY, these changes define a region distinct from the active site. A single amino-acid substitution within this binding region on CheY, alanine to valine at position 103, significantly decreases the affinity of CheY for CheA. The binding face described by these changes partially overlaps a flagellar switch binding surface previously defined by mutagenesis.