Structure of biomimetic casein micelles: Critical tests of the hydrophobic colloid and multivalent-binding models using recombinant deuterated and phosphorylated β-casein

Milk contains high concentrations of amyloidogenic casein proteins and is supersaturated with respect to crystalline calcium phosphates such as apatite. Nevertheless, the mammary gland normally remains unmineralized and free of amyloid. Unlike κ-casein, β- and αS-caseins are highly effective mineral chaperones that prevent ectopic and pathological calcification of the mammary gland. Milk invariably contains a mixture of two to five different caseins that act on each other as molecular chaperones. Instead of forming amyloid fibrils, several thousand caseins and hundreds of nanoclusters of amorphous calcium phosphate combine to form fuzzy complexes called casein micelles. To understand the biological functions of the casein micelle its structure needs to be understood better than at present. The location in micelles of the highly amyloidogenic κ-casein is disputed. In traditional hydrophobic colloid models, it, alone, forms a stabilizing surface coat that also determines the average size of the micelles. In the recent multivalent-binding model, κ-casein is present throughout the micelle, in intimate contact with the other caseins. To discriminate between these models, a range of biomimetic micelles was prepared using a fixed concentration of the mineral chaperone β-casein and nanoclusters of calcium phosphate, with variable concentrations of κ-casein. A biomimetic micelle was also prepared using a highly deuterated and in vivo phosphorylated recombinant β-casein with calcium phosphate and unlabelled κ-casein. Neutron and X-ray scattering experiments revealed that κ-casein is distributed throughout the micelle, in quantitative agreement with the multivalent-binding model but contrary to the hydrophobic colloid models.

NMR characterization of an assembling RHIM (RIP homotypic interaction motif) amyloid reveals a cryptic region for self-recognition

The RIP homotypic interaction motif (RHIM) is an essential protein motif in inflammatory signaling and certain cell death pathways. RHIM signaling occurs following the assembly of functional amyloids, and while the structural biology of such higher-order RHIM complexes has started to emerge, the conformations and dynamics of nonassembled RHIMs remain unknown. Here, using solution NMR spectroscopy, we report the characterization of the monomeric form of the RHIM in receptor-interacting protein kinase 3 (RIPK3), a fundamental protein in human immunity. Our results establish that the RHIM of RIPK3 is an intrinsically disordered protein motif, contrary to prediction, and that exchange dynamics between free monomers and amyloid-bound RIPK3 monomers involve a 20-residue stretch outside the RHIM that is not incorporated within the structured cores of the RIPK3 assemblies determined by cryo-EM or solid-state NMR. Thus, our findings expand on the structural characterization of RHIM-containing proteins, specifically highlighting conformational dynamics involved in assembly processes.

Structural basis underlying the synergism of NADase and SLO during group A Streptococcus infection

Group A Streptococcus (GAS) is a strict human pathogen possessing a unique pathogenic trait that utilizes the cooperative activity of NAD+-glycohydrolase (NADase) and Streptolysin O (SLO) to enhance its virulence. How NADase interacts with SLO to synergistically promote GAS cytotoxicity and intracellular survival is a long-standing question. Here, the structure and dynamic nature of the NADase/SLO complex are elucidated by X-ray crystallography and small-angle scattering, illustrating atomic details of the complex interface and functionally relevant conformations. Structure-guided studies reveal a salt-bridge interaction between NADase and SLO is important to cytotoxicity and resistance to phagocytic killing during GAS infection. Furthermore, the biological significance of the NADase/SLO complex in GAS virulence is demonstrated in a murine infection model. Overall, this work delivers the structure-functional relationship of the NADase/SLO complex and pinpoints the key interacting residues that are central to the coordinated actions of NADase and SLO in the pathogenesis of GAS infection.

Identification and characterization of two drug-like fragments that bind to the same cryptic binding pocket of Burkholderia pseudomallei DsbA

Disulfide-bond-forming proteins (Dsbs) play a crucial role in the pathogenicity of many Gram-negative bacteria. Disulfide-bond-forming protein A (DsbA) catalyzes the formation of the disulfide bonds necessary for the activity and stability of multiple substrate proteins, including many virulence factors. Hence, DsbA is an attractive target for the development of new drugs to combat bacterial infections. Here, two fragments, bromophenoxy propanamide (1) and 4-methoxy-N-phenylbenzenesulfonamide (2), were identified that bind to DsbA from the pathogenic bacterium Burkholderia pseudomallei, the causative agent of melioidosis. The crystal structures of oxidized B. pseudomallei DsbA (termed BpsDsbA) co-crystallized with 1 or 2 show that both fragments bind to a hydrophobic pocket that is formed by a change in the side-chain orientation of Tyr110. This conformational change opens a `cryptic' pocket that is not evident in the apoprotein structure. This binding location was supported by 2D-NMR studies, which identified a chemical shift perturbation of the Tyr110 backbone amide resonance of more than 0.05 p.p.m. upon the addition of 2 mM fragment 1 and of more than 0.04 p.p.m. upon the addition of 1 mM fragment 2. Although binding was detected by both X-ray crystallography and NMR, the binding affinity (Kd) for both fragments was low (above 2 mM), suggesting weak interactions with BpsDsbA. This conclusion is also supported by the crystal structure models, which ascribe partial occupancy to the ligands in the cryptic binding pocket. Small fragments such as 1 and 2 are not expected to have a high energetic binding affinity due to their relatively small surface area and the few functional groups that are available for intermolecular interactions. However, their simplicity makes them ideal for functionalization and optimization. The identification of the binding sites of 1 and 2 to BpsDsbA could provide a starting point for the development of more potent novel antimicrobial compounds that target DsbA and bacterial virulence.

NMR fragment screening reveals a novel small molecule binding site near the catalytic surface of the disulfide-dithiol oxidoreductase enzyme DsbA from Burkholderia pseudomallei

The presence of suitable cavities or pockets on protein structures is a general criterion for a therapeutic target protein to be classified as 'druggable'. Many disease-related proteins that function solely through protein-protein interactions lack such pockets, making development of inhibitors by traditional small-molecule structure-based design methods much more challenging. The 22 kDa bacterial thiol oxidoreductase enzyme, DsbA, from the gram-negative bacterium Burkholderia pseudomallei (BpsDsbA) is an example of one such target. The crystal structure of oxidized BpsDsbA lacks well-defined surface pockets. BpsDsbA is required for the correct folding of numerous virulence factors in B. pseudomallei, and genetic deletion of dsbA significantly attenuates B. pseudomallei virulence in murine infection models. Therefore, BpsDsbA is potentially an attractive drug target. Herein we report the identification of a small molecule binding site adjacent to the catalytic site of oxidized BpsDsbA. ¹H^N CPMG relaxation dispersion NMR measurements suggest that the binding site is formed transiently through protein dynamics. Using fragment-based screening, we identified a small molecule that binds at this site with an estimated affinity of K_D ~ 500 µM. This fragment inhibits BpsDsbA enzymatic activity in vitro. The binding mode of this molecule has been characterized by NMR data-driven docking using HADDOCK. These data provide a starting point towards the design of more potent small molecule inhibitors of BpsDsbA.

Transient antibody-antigen interactions mediate the strain-specific recognition of a conserved malaria epitope

Transient interactions in which binding partners retain substantial conformational disorder play an essential role in regulating biological networks, challenging the expectation that specificity demands structurally defined and unambiguous molecular interactions. The monoclonal antibody 6D8 recognises a completely conserved continuous nine-residue epitope within the intrinsically disordered malaria antigen, MSP2, yet it has different affinities for the two allelic forms of this antigen. NMR chemical shift perturbations, relaxation rates and paramagnetic relaxation enhancements reveal the presence of transient interactions involving polymorphic residues immediately C-terminal to the structurally defined epitope. A combination of these experimental data with molecular dynamics simulations shows clearly that the polymorphic C-terminal extension engages in multiple transient interactions distributed across much of the accessible antibody surface. These interactions are determined more by topographical features of the antibody surface than by sequence-specific interactions. Thus, specificity arises as a consequence of subtle differences in what are highly dynamic and essentially non-specific interactions.

Structure and Characterisation of a Key Epitope in the Conserved C-Terminal Domain of the Malaria Vaccine Candidate MSP2

Merozoite surface protein 2 (MSP2) is an intrinsically disordered antigen that is abundant on the surface of the malaria parasite Plasmodium falciparum. The two allelic families of MSP2, 3D7 and FC27, differ in their central variable regions, which are flanked by highly conserved C-terminal and N-terminal regions. In a vaccine trial, full-length 3D7 MSP2 induced a strain-specific protective immune response despite the detectable presence of conserved region antibodies. This work focuses on the conserved C-terminal region of MSP2, which includes the only disulphide bond in the protein and encompasses key epitopes recognised by the mouse monoclonal antibodies 4D11 and 9H4. Although the 4D11 and 9H4 epitopes are overlapping, immunofluorescence assays have shown that the mouse monoclonal antibody 4D11 binds to MSP2 on the merozoite surface with a much stronger signal than 9H4. Understanding the structural basis for this antigenic difference between these antibodies will help direct the design of a broad-spectrum and MSP2-based malaria vaccine. 4D11 and 9H4 were reengineered into antibody fragments [variable region fragment (Fv) and single-chain Fv (scFv)] and were validated as suitable models for their full-sized IgG counterparts by surface plasmon resonance and isothermal titration calorimetry. An alanine scan of the 13-residue epitope 3D7-MSP2_207-222 identified the minimal binding epitope of 4D11 and the key residues involved in binding. A 2.2-Å crystal structure of 4D11 Fv bound to the eight-residue epitope NKENCGAA provided valuable insight into the possible conformation of the C-terminal region of MSP2 on the parasite. This work underpins continued efforts to optimise recombinant MSP2 constructs for evaluation as potential vaccine candidates.

Solid-state NMR spectroscopy of functional amyloid from a fungal hydrophobin: a well-ordered β-sheet core amidst structural heterogeneity

GrEASy fibrils: Hydrophobins are fungal proteins that assemble into an amphipathic fibrillar monolayer with amyloid properties and a hydrophobic face as water-resistant as Teflon. Solid-state NMR studies on EAS hydrophobin fibrils reveal direct evidence of a partial molecular rearrangement on assembly and an ordered β-sheet-rich core in the context of a whole protein in this functional amyloid.

Production and use of deuterated polyhydroxyoctanoate in structural studies of PHO inclusions☆

This work reports on the biosynthesis of polyhydroxyalkanoates with medium chain length alkyl substituents in the side chain by Pseudomonas oleovorans using hydrogenated and deuterated substrates. These investigations aimed to obtain polyhydroxyalkanoates with varying degrees of deuterium substitution, and establish whether they are suitable analogues for structural investigation. In order to understand the formation and structure of inclusions in their native state, whole inclusions were isolated from microbial cells and were analysed using Small Angle Neutron Scattering. A contrast variation study was conducted on hydrogenated and deuterated inclusions of polyhydroxyoctanoate, as well as inclusions resulting from co-feeding or sequentially feeding different precursors. The data indicated a core/shell structure resulting from feeding hydrogenated followed by perdeuterated PHO precursor, and demonstrated the utility of this analysis for characterising chemically similar systems.

The effect of pH on the uptake and toxicity of copper and zinc in a tropical freshwater alga (Chlorella sp.)

Copper and zinc toxicity to the freshwater alga Chlorella sp. was determined at a range of pH values (5.5-8.0) in a synthetic softwater (hardness 40-48 mg CaCO(3)/L). The effects of the metals on algal growth (cell division) rate were determined after 48-h exposure at pH 5.5, 6.0, 6.5, 7.0, 7.5, and 8.0. The toxicity of both metals was pH dependent. As pH decreased from 8.0 to 5.5, the copper concentration required to inhibit the algal growth rate by 50% (IC50) increased from 1.0 to 19 microg/L. For zinc, the IC50 increased from 52 to 2,700 microg/L over the same pH range. Changes in solution speciation alone did not explain the increased toxicity observed as the pH increased. Modelled Cu(2+) and Zn(2+) concentrations decreased with increasing pH, whereas toxicity was observed to increase. Measurements of extracellular (cell-bound) metal concentrations support the biotic ligand model (BLM) theory of competition between protons (H(+)) and metals for binding sites at the algal cell surface. Higher extracellular metal concentrations were observed at high pH, indicating reduced competition. Independent of pH, both extracellular and intracellular copper were directly related to growth inhibition in Chlorella sp., whereas zinc toxicity was related to cell-bound zinc only. These findings suggest that the algal cell surface may be considered as the biotic ligand in further development of a chronic BLM with microalgae. Conditional binding constants (log K) were determined experimentally (using measured intracellular metal concentrations) and theoretically (using concentration-response curves) for copper and zinc for Chlorella sp. at selected pH values. Excellent agreement was found indicating the possibility of using concentration-response data to estimate conditional metal-cell binding constants.

Hardness corrections for copper are inappropriate for protecting sensitive freshwater biota

Toxicity testing using a freshwater alga (Chlorella sp.), a bacterium (Erwinnia sp.) and a cladoceran (Ceriodaphnia cf. dubia) exposed to copper in synthetic and natural freshwaters of varying hardness (44-375 mg CaCO3/l), with constant alkalinity, pH and dissolved organic carbon concentration, demonstrated negligible hardness effects in the pH range 6.1-7.8. Therefore, the use of a generic hardness-correction algorithm, developed as part of national water quality guidelines for protecting freshwater biota, is not recommended for assessing the toxicity of copper to these, and other, sensitive freshwater species. Use of the algorithm for these sensitive species will be underprotective because the calculated concentrations of copper in water that cause a toxic effect will be higher.

Toward a biotic ligand model for freshwater green algae: surface-bound and internal copper are better predictors of toxicity than free Cu2+-ion activity when pH is varied

The freshwater green microalgae Chlorella sp. and Pseudokirchneriella subcapitata (P. subcapitata) were chronically (48 and 72 h, respectively) exposed to copper at various pH levels, i.e., pH 6-7.5 and pH 5.9-8.5, respectively. Concentrations resulting in 50% inhibition of exponential growth rate (EC50) were determined as dissolved Cu, estimated chemical activity of the free Cu2+ ion (as pCu = - log{Cu2+ activity as molarity}), and as external (surface-bound) Cu and internal Cu in the algal cells. With increasing pH, EC50dissolved decreased from 30 to 1.1 microg of Cu L(-1) for Chlorella sp. and from 46 to 18 microg of Cu L(-1) for P. subcapitata. The pH effect on copper toxicity was even more obvious when expressed as Cu2+ activity. The EC50pCu increased on average 1.4 pCu unit per pH unit for Chlorella sp. and 1.1 pCu unit per pH unit for P. subcapitata, thus indicating a marked increase of Cu2+ toxicity at higher pH (more than 1 order of magnitude per pH unit). In contrast, it was found that EC50 values expressed as surface bound or external copper (EC50external) and as internal copper (EC50internal) did not vary substantially when pH was increased. External Cu was operationally defined as the Cu fraction removable from the algal cell by short-term contact with ethylenediaminetetraacetic acid; internal copper was defined as the nonremovable fraction. For Chlorella sp. the EC50external varied between 5 and 10 fg of Cu/ cell (factor of 2 difference) and the EC50internal between 25 and 40 fg of Cu/cell (factor of 1.6 difference). For P. subcapitata the EC50external varied between 10 and 28 fg of Cu/cell (factor of 2.8 difference) and the EC50internal between 42 and 71 fg of Cu/cell (factor of 1.7 difference). Because the observed variation in EC50external and EC50internal is much less than the variation in EC50Cu2+, it is concluded that both external and internal copper are better predictors of copper toxicity than Cu2+ when pH is varied. From the perspective of toxicity modeling, this observation is the first step toward considering the use of the cell surface as the algal biotic ligand for Cu in a similar way as fish gills fulfill this role in the biotic ligand model for predicting metal toxicity to fish species.