Rational Design of Monodispersed Mutants of Proteins by Identifying Aggregation Contact Sites Using Solubilizing Agents.

Suppression of protein aggregation is a subject of growing importance in the treatment of protein aggregation diseases, an urgent worldwide human health problem, and the production of therapeutic proteins, such as antibody drugs. We previously reported a method to identify compounds that suppress aggregation, based on screening using multiple terminal deletion mutants. We now present a method to determine the aggregation contact sites of proteins, using such solubilizing compounds, to design monodispersed mutants. We applied this strategy to the chemokine receptor-binding domain (CRBD) of FROUNT, which binds to the membrane-proximal C-terminal intracellular region of CCR2. Initially, the backbone NMR signals were assigned to a certain extent by available methods, and the putative locations of five α-helices were identified. Based on NMR chemical shift perturbations upon varying the protein concentrations, the first and third helices were found to contain the aggregation contact sites. The two helices are amphiphilic, and based on an NMR titration with 1,6-hexanediol, a CRBD solubilizing compound, the contact sites were identified as the hydrophobic patches located on the hydrophilic sides of the two helices. Subsequently, we designed multiple mutants targeting amino acid residues on the contact sites. Based on their NMR spectra, a doubly mutated CRBD (L538E/P612S) was selected from the designed mutants, and its monodispersed nature was confirmed by other biophysical methods. We then assessed the CCR2-binding activities of the mutants. Our method is useful for the protein structural analyses, the treatment of protein aggregation diseases, and the improvement of therapeutic proteins.

1H, 13C and 15N resonance assignments for a chemokine receptor-binding domain of FROUNT, a cytoplasmic regulator of chemotaxis.

FROUNT is a cytoplasmic protein that interacts with the membrane-proximal C-terminal regions (Pro-Cs) of the CCR2 and CCR5 chemokine receptors. The interactions between FROUNT and the chemokine receptors play an important role in the migration of inflammatory immune cells. Therefore, FROUNT is a potential drug target for inflammatory diseases. However, the structural basis of the interactions between FROUNT and the chemokine receptors remains to be elucidated. We previously identified the C-terminal region (residues 532-656) of FROUNT as the structural domain responsible for the Pro-C binding, referred to as the chemokine receptor-binding domain (CRBD), and then constructed its mutant, bearing L538E/P612S mutations, with improved NMR spectral quality, referred to as CRBD_LEPS. We now report the main-chain and side-chain 1H, 13C, and 15N resonance assignments of CRBD_LEPS. The NMR signals of CRBD_LEPS were well dispersed and their intensities were uniform on the 1H-15N HSQC spectrum, and thus almost all of the main-chain and side-chain resonances were assigned. This assignment information provides the foundation for NMR studies of the three-dimensional structure of CRBD_LEPS in solution and its interactions with chemokine receptors.

Efficient identification of compounds suppressing protein precipitation via solvent screening using serial deletion mutants of the target protein.

The control of protein solubility is a subject of broad interest. Although several solvent screening methods are available to search for compounds that enhance protein solubilization, their performance is influenced by the intrinsic solubility of the tested protein. We now present a method for screening solubilizing compounds, using an array of N- or C-terminal deletion mutants of the protein. A key behind this approach is that such terminal deletions of the protein affect its aggregation propensity. The solubilization activities of trial solvents are individually assessed, based on the number of solubilized mutants. The solubilizing compounds are then identified from the screened solvents. In this study, the C-terminal chemokine receptor-binding region of the cytoplasmic protein, FROUNT (FNT-C), which mediates intracellular signals leading to leukocyte migration, was subjected to the multicomponent screening. In total, 192 solution conditions were tested, using eight terminal deletion mutants of FNT-C. We identified five solvent conditions that solubilized four or five mutants of FNT-C, and the compounds in the screened solvents were then, respectively, assessed in terms of their solubilization ability. The best compound for solubilizing FNT-C was 1,6-hexanediol. Indeed, 1,6-hexanediol bound to FNT-C and suppressed its precipitation, as showed by NMR and dynamic light scattering analyses.

Structural basis for the binding of the membrane-proximal C-terminal region of chemokine receptor CCR2 with the cytosolic regulator FROUNT.

The membrane-proximal C-terminal region (Pro-C) is important for the regulation of G-protein-coupled receptors (GPCRs), but the binding of the Pro-C region to a cytosolic regulator has not been structurally analyzed. The chemokine receptor CCR2 is a member of the GPCR superfamily, and the Pro-C region of CCR2 binds to the cytosolic regulator FROUNT. Studying the interaction between CCR2 Pro-C and FROUNT at an atomic level provides a basis for understanding the signal transduction mechanism via GPCRs. NOE-based NMR experiments showed that, when bound to FROUNT, CCR2 Pro-C adopted a helical conformation, as well as when embedded in dodecylphosphocholine micelles. A comparison of two types of cross-saturation-based NMR experiments, applied to a three-component mixture of Pro-C, FROUNT and micelles or a two-component mixture of Pro-C and micelles, revealed that the hydrophobic binding surface on Pro-C for FROUNT mostly overlapped with the binding site for micelles, suggesting competitive binding of Pro-C between FROUNT and micelles. Leu316 was important for both FROUNT and micelle binding. Phe319 was newly identified to be crucial for FROUNT binding, by NMR and mutational analyses. The association and dissociation rates of CCR2 Pro-C for lipid bilayer biomembranes were faster than those for FROUNT. We previously reported that FROUNT binding to CCR2 is detectable even in unstimulated cells and increases in response to chemokine stimulation. Taken together, these results support a model of CCR2 equilibrium: chemokine binding changes the conformational equilibrium of CCR2 toward the active state, and Pro-C switches its binding partner from the membrane to FROUNT.