Hydrogen/Deuterium Exchange and Nuclear Magnetic Resonance Spectroscopy Reveal Dynamic Allostery on Multiple Time Scales in the Serine Protease Thrombin

The Light Chain Allosterically Enhances the Protease Activity of Murine Urokinase-Type Plasminogen Activator

The active form of the murine urokinase-type plasminogen activator (muPA) is formed by a 27-residue disordered light chain connecting the amino-terminal fragment (ATF) with the serine protease domain. The two chains are tethered by a disulfide bond between C1CT in the disordered light chain and C122CT in the protease domain. Previous work showed that the presence of the disordered light chain affected the inhibition of the protease domain by antibodies. Here we show that the disordered light chain induced a 3.7-fold increase in kcat of the protease domain of muPA. In addition, hydrogen-deuterium exchange mass spectrometry (HDX-MS) and accelerated molecular dynamics (AMD) were performed to identify the interactions between the disordered light chain and the protease domain. HDX-MS revealed that the light chain is contacting the 110s, the turn between the β10- and β11-strand, and the β7-strand. A reduction in deuterium uptake was also observed in the activation loop, the 140s loop and the 220s loop, which forms the S1-specificty pocket where the substrate binds. These loops are further away from where the light chain seems to be interacting with the protease domain. Our results suggest that the light chain most likely increases the activity of muPA by allosterically favoring conformations in which the specificity pocket is formed. We propose a model by which the allostery would be transmitted through the β-strands of the β-barrels to the loops on the other side of the protease domain.

Dynamics, allostery, and stabilities of whole virus particles by amide hydrogen/deuterium exchange mass spectrometry (HDXMS)

X-ray crystallography and cryo-electron microscopy have enabled the determination of structures of numerous viruses at high resolution and have greatly advanced the field of structural virology. These structures represent only a subset of snapshot end-state conformations, without describing all conformational transitions that virus particles undergo. Allostery plays a critical role in relaying the effects of varied perturbations both on the surface through environmental changes and protein (receptor/antibody) interactions into the genomic core of the virus. Correspondingly, allostery carries implications for communicating changes in genome packaging to the overall stability of the virus particle. Amide hydrogen/deuterium exchange mass spectrometry (HDXMS) of whole viruses is a powerful probe for uncovering virus allostery. Here we critically discuss advancements in understanding virus dynamics by HDXMS with single particle cryo-EM and computational approaches.

Structure of Human Serum Albumin at a Foam Surface

Proteins can be adsorbed on the air-water interface (AWI), and the structural changes in proteins at the AWI are closely related to the foaming properties of foods and beverages. However, how these structural changes in proteins at the AWI occur is not well understood. We developed a method for the structural assessment of proteins in the foam state using hydrogen/deuterium exchange mass spectrometry. Adsorption sites and structural changes in human serum albumin (HSA) were identified in situ at the peptide-level resolution. The N-terminus and the loop (E492-T506), which contains hydrophobic amino acids, were identified as adsorption sites. Both the structural flexibility and hydrophobicity were considered to be critical factors for the adsorption of HSA at the AWI. Structural changes in HSA were observed after more than one minute of foaming and were spread widely throughout the structure. These structural changes at the foam AWI were reversible.

Identification of the Thermal Activation Network in Human 15-Lipoxygenase-2: Divergence from Plant Orthologs and Its Relationship to Hydrogen Tunneling Activation Barriers

The oxidation of polyunsaturated fatty acids by lipoxygenases (LOXs) is initiated by a C-H cleavage step in which the hydrogen atom is transferred quantum mechanically (i.e., via tunneling). In these reactions, protein thermal motions facilitate the conversion of ground-state enzyme-substrate complexes to tunneling-ready configurations and are thus important for transferring energy from the solvent to the active site for the activation of catalysis. In this report, we employed temperature-dependent hydrogen-deuterium exchange mass spectrometry (TDHDX-MS) to identify catalytically linked, thermally activated peptides in a representative animal LOX, human epithelial 15-LOX-2. TDHDX-MS of wild-type 15-LOX-2 was compared to two active site mutations that retain structural stability but have increased activation energies (Ea) of catalysis. The Ea value of one variant, V427L, is implicated to arise from suboptimal substrate positioning by increased active-site side chain rotamer dynamics, as determined by X-ray crystallography and ensemble refinement. The resolved thermal network from the comparative Eas of TDHDX-MS between wild-type and V426A is localized along the front face of the 15-LOX-2 catalytic domain. The network contains a clustering of isoleucine, leucine, and valine side chains within the helical peptides. This thermal network of 15-LOX-2 is different in location, area, and backbone structure compared to a model plant lipoxygenase from soybean that exhibits a low Ea value of catalysis compared to the human ortholog. The presented data provide insights into the divergence of thermally activated protein motions in plant and animal LOXs and their relationships to the enthalpic barriers for facilitating hydrogen tunneling.

Exploring Residue-Level Interactions between the Biofilm-Driving R2ab Protein and Polystyrene Nanoparticles

In biological systems, proteins can bind to nanoparticles to form a "corona" of adsorbed molecules. The nanoparticle corona is of significant interest because it impacts an organism's response to a nanomaterial. Understanding the corona requires knowledge of protein structure, orientation, and dynamics at the surface. A residue-level mapping of protein behavior on nanoparticle surfaces is needed, but this mapping is difficult to obtain with traditional approaches. Here, we have investigated the interaction between R2ab and polystyrene nanoparticles (PSNPs) at the level of individual residues. R2ab is a bacterial surface protein from Staphylococcus epidermidis and is known to interact strongly with polystyrene, leading to biofilm formation. We have used mass spectrometry after lysine methylation and hydrogen-deuterium exchange (HDX) NMR spectroscopy to understand how the R2ab protein interacts with PSNPs of different sizes. Lysine methylation experiments reveal subtle but statistically significant changes in methylation patterns in the presence of PSNPs, indicating altered protein surface accessibility. HDX rates become slower overall in the presence of PSNPs. However, some regions of the R2ab protein exhibit faster than average exchange rates in the presence of PSNPs, while others are slower than the average behavior, suggesting conformational changes upon binding. HDX rates and methylation ratios support a recently proposed "adsorbotope" model for PSNPs, wherein adsorbed proteins consist of unfolded anchor points interspersed with partially structured regions. Our data also highlight the challenges of characterizing complex protein-nanoparticle interactions using these techniques, such as fast exchange rates. While providing insights into how R2ab adsorbs onto PSNP surfaces, this research emphasizes the need for advanced methods to comprehend residue-level interactions in the nanoparticle corona.

From haemadin to haemanorm: Synthesis and characterization of full-length haemadin from the leech Haemadipsa sylvestris and of a novel bivalent, highly potent thrombin inhibitor (haemanorm)

Hirudin from Hirudo medicinalis is a bivalent α-Thrombin (αT) inhibitor, targeting the enzyme active site and exosite-I, and is currently used in anticoagulant therapy along with its simplified analogue hirulog. Haemadin, a small protein (57 amino acids) isolated from the land-living leech Haemadipsa sylvestris, selectively inhibits αT with a potency identical to that of recombinant hirudin (KI  = 0.2 pM), with which it shares a common disulfide topology and overall fold. At variance with hirudin, haemadin targets exosite-II and therefore (besides the free protease) it also blocks thrombomodulin-bound αT without inhibiting the active intermediate meizothrombin, thus offering potential advantages over hirudin. Here, we produced in reasonably high yields and pharmaceutical purity (>98%) wild-type haemadin and the oxidation resistant Met5 → nor-Leucine analogue, both inhibiting αT with a KI of 0.2 pM. Thereafter, we used site-directed mutagenesis, spectroscopic, ligand-displacement, and Hydrogen/Deuterium Exchange-Mass Spectrometry techniques to map the αT regions relevant for the interaction with full-length haemadin and with the synthetic N- and C-terminal peptides Haem(1-10) and Haem(45-57). Haem(1-10) competitively binds to/inhibits αT active site (KI  = 1.9 μM) and its potency was enhanced by 10-fold after Phe3 → β-Naphthylalanine exchange. Conversely to full-length haemadin, haem(45-57) displays intrinsic affinity for exosite-I (KD  = 1.6 μM). Hence, we synthesized a peptide in which the sequences 1-9 and 45-57 were joined together through a 3-Glycine spacer to yield haemanorm, a highly potent (KI  = 0.8 nM) inhibitor targeting αT active site and exosite-I. Haemanorm can be regarded as a novel class of hirulog-like αT inhibitors with potential pharmacological applications.

The autoactivation of human single-chain urokinase-type plasminogen activator (uPA)

Most serine proteases are synthesized as inactive zymogens that are activated by cleavage by another protease in a tightly regulated mechanism. The urokinase-type plasminogen activator (uPA) and plasmin cleave and activate each other, constituting a positive feedback loop. How this mutual activation cycle begins has remained a mystery. We used hydrogen deuterium exchange mass spectrometry to characterize the dynamic differences between the inactive single-chain uPA (scuPA) and its active form two-chain uPA (tcuPA). The results show that the C-terminal β-barrel and the area around the new N terminus have significantly reduced dynamics in tcuPA as compared with scuPA. We also show that the zymogen scuPA is inactive but can, upon storage, become active in the absence of external proteases. In addition to plasmin, the tcuPA can activate scuPA by cleavage at K158, a process called autoactivation. Unexpectedly, tcuPA can cleave at position 158 even when this site is mutated. TcuPA can also cleave scuPA after K135 or K136 in the disordered linker, which generates the soluble protease domain of uPA. Plasmin cleaves scuPA exclusively after K158 and at a faster rate than tcuPA. We propose a mechanism by which the uPA receptor dimerization could promote autoactivation of scuPA on cell surfaces. These results resolve long-standing controversies in the literature surrounding the mechanism of uPA activation.

Exploring the Residue-Level Interactions between the R2ab Protein and Polystyrene Nanoparticles

In biological systems, proteins can bind to nanoparticles to form a "corona" of adsorbed molecules. The nanoparticle corona is of high interest because it impacts the organism's response to the nanomaterial. Understanding the corona requires knowledge of protein structure, orientation, and dynamics at the surface. Ultimately, a residue-level mapping of protein behavior on nanoparticle surfaces is needed, but this mapping is difficult to obtain with traditional approaches. Here, we have investigated the interaction between R2ab and polystyrene nanoparticles (PSNPs) at the level of individual residues. R2ab is a bacterial surface protein from Staphylococcus epidermidis and is known to interact strongly with polystyrene, leading to biofilm formation. We have used mass spectrometry after lysine methylation and hydrogen-deuterium exchange (HDX) NMR spectroscopy to understand how the R2ab protein interacts with PSNPs of different sizes. Through lysine methylation, we observe subtle but statistically significant changes in methylation patterns in the presence of PSNPs, indicating altered protein surface accessibility. HDX measurements reveal that certain regions of the R2ab protein undergo faster exchange rates in the presence of PSNPs, suggesting conformational changes upon binding. Both results support a recently proposed "adsorbotope" model, wherein adsorbed proteins consist of unfolded anchor points interspersed with regions of partial structure. Our data also highlight the challenges of characterizing complex protein-nanoparticle interactions using these techniques, such as fast exchange rates. While providing insights into how proteins respond to nanoparticle surfaces, this research emphasizes the need for advanced methods to comprehend these intricate interactions fully at the residue level.

AlphaFold, allosteric, and orthosteric drug discovery: Ways forward

Drug discovery is arguably a highly challenging and significant interdisciplinary aim. The stunning success of the artificial intelligence-powered AlphaFold, whose latest version is buttressed by an innovative machine-learning approach that integrates physical and biological knowledge about protein structures, raised drug discovery hopes that unsurprisingly, have not come to bear. Even though accurate, the models are rigid, including the drug pockets. AlphaFold's mixed performance poses the question of how its power can be harnessed in drug discovery. Here we discuss possible ways of going forward wielding its strengths, while bearing in mind what AlphaFold can and cannot do. For kinases and receptors, an input enriched in active (ON) state models can better AlphaFold's chance of rational drug design success.

Capturing the Dynamic Conformational Changes of Human Isocitrate Dehydrogenase 1 (IDH1) upon Ligand and Metal Binding Using Hydrogen-Deuterium Exchange Mass Spectrometry

Human isocitrate dehydrogenase 1 (IDH1) is a highly conserved metabolic enzyme that catalyzes the interconversion of isocitrate and α-ketoglutarate. Kinetic and structural studies with IDH1 have revealed evidence of striking conformational changes that occur upon binding of its substrates, isocitrate and NADP⁺, and its catalytic metal cation. Here, we used hydrogen-deuterium exchange mass spectrometry (HDX-MS) to build a comprehensive map of the dynamic conformational changes experienced by IDH1 upon ligand binding. IDH1 proved well-suited for HDX-MS analysis, allowing us to capture profound changes in solvent accessibility at substrate binding sites and at a known regulatory region, as well as at more distant local subdomains that appear to support closure of this protein into its active conformation. HDX-MS analysis suggested that IDH1 is primarily purified with NADP(H) bound in the absence of its metal cation. Subsequent metal cation binding, even in the absence of isocitrate, was critical for driving large conformational changes. WT IDH1 folded into its fully closed conformation only when the full complement of substrates and metal was present. Finally, we show evidence supporting a previously hypothesized partially open conformation that forms prior to the catalytically active state, and we propose this conformation is driven by isocitrate binding in the absence of metal.

Ligand-specific changes in conformational flexibility mediate long-range allostery in the lac repressor

Biological regulation ubiquitously depends on protein allostery, but the regulatory mechanisms are incompletely understood, especially in proteins that undergo ligand-induced allostery with few structural changes. Here we used hydrogen-deuterium exchange with mass spectrometry (HDX/MS) to map allosteric effects in a paradigm ligand-responsive transcription factor, the lac repressor (LacI), in different functional states (apo, or bound to inducer, anti-inducer, and/or DNA). Although X-ray crystal structures of the LacI core domain in these states are nearly indistinguishable, HDX/MS experiments reveal widespread differences in flexibility. We integrate these results with modeling of protein-ligand-solvent interactions to propose a revised model for allostery in LacI, where ligand binding allosterically shifts the conformational ensemble as a result of distinct changes in the rigidity of secondary structures in the different states. Our model provides a mechanistic basis for the altered function of distal mutations. More generally, our approach provides a platform for characterizing and engineering protein allostery.

Allostery Frustrates the Experimentalist

Proteins interact with other proteins, with nucleic acids, lipids, carbohydrates and various small molecules in the living cell. These interactions have been quantified and structurally characterized in numerous studies such that we today have a comprehensive picture of protein structure and function. However, proteins are dynamic and even folded proteins are likely more heterogeneous than they appear in most descriptions. One property of proteins that relies on dynamics and heterogeneity is allostery, the ability of a protein to change structure and function upon ligand binding to an allosteric site. Over the last decades the concept of allostery was broadened to embrace all types of long-range interactions across a protein including purely entropic changes without a conformational change in single protein domains. But with this re-definition came a problem: How do we measure allostery? In this opinion, we discuss some caveats arising from the quantitative description of single-domain allostery from an experimental perspective and how the limitations cannot be separated from the definition of allostery per se. Furthermore, we attempt to tie together allostery with the concept of frustration in an effort to investigate the links between these two complex, and yet general, properties of proteins. We arrive at the conclusion that the sensitivity to perturbation of allosteric networks in single protein domains is too large for the networks to be of significant biological relevance.

Biochemical and hydrogen-deuterium exchange studies of the single nucleotide polymorphism Y649C in human platelet 12-lipoxygenase linked to a bleeding disorder

Human platelet 12-lipoxygenase (h12-LOX) is responsible for the formation of oxylipin products that play an important role in platelet aggregation. Single nucleotide polymorphisms (SNPs) of h12-LOX have been implicated in several diseases. In this study, we investigate the structural, dynamical, and functional impact of a h12-LOX SNP that generates a tyrosine-to-cysteine mutation at a buried site (Y649C h12-LOX) and was previously ascribed with reduced levels of 12(S)-hydroxyeicosatetraenoic acid (12S-HETE) production in isolated platelets. Herein, in vitro Michaelis-Menten kinetics show reduced catalytic rates for Y649C compared to WT h12-LOX at physiological or lower temperatures. Both proteins exhibited similar melting temperatures, metal content, and oligomerization state. Liposome binding for both proteins was also dependent upon the presence of calcium, temperature, and liposome composition; however, the Y649C variant was found to have lowered binding capacity to liposomes compared to WT at physiological temperatures. Further, hydrogen-deuterium exchange mass spectrometry (HDX-MS) experiments revealed a regional defined enhancement in the peptide mobility caused by the mutation. This increased instability for the mutation stemmed from a change in an interaction with an arched helix that lines the substrate binding site, located ≥15 Å from the mutation site. Finally, differential scanning calorimetry demonstrated a reduced protein (un)folding enthalpy, consistent with the HDX results. Taken together, these results demonstrate remarkable similarity between the mutant and WT h12-LOX, and yet, subtle changes in activity, membrane affinity and protein stability may be responsible for the significant physiological changes that the Y649C SNP manifests in platelet biology.

An intrinsically disordered transcription activation domain increases the DNA binding affinity and reduces the specificity of NFκB p50/RelA

Many transcription factors contain intrinsically disordered transcription activation domains (TADs), which mediate interactions with coactivators to activate transcription. Historically, DNA-binding domains and TADs have been considered as modular units, but recent studies have shown that TADs can influence DNA binding. Whether these results can be generalized to more TADs is not clear. Here, we biophysically characterized the NFκB p50/RelA heterodimer including the RelA TAD and investigated the TAD's influence on NFκB-DNA interactions. In solution, we show the RelA TAD is disordered but compact, with helical tendency in two regions that interact with coactivators. We determined that the presence of the TAD increased the stoichiometry of NFκB-DNA complexes containing promoter DNA sequences with tandem κB recognition motifs by promoting the binding of NFκB dimers in excess of the number of κB sites. In addition, we measured the binding affinity of p50/RelA for DNA containing tandem κB sites and single κB sites. While the presence of the TAD enhanced the binding affinity of p50/RelA for all κB sequences tested, it also increased the affinity for nonspecific DNA sequences by over 10-fold, leading to an overall decrease in specificity for κB DNA sequences. In contrast, previous studies have generally reported that TADs decrease DNA-binding affinity and increase sequence specificity. Our results reveal a novel function of the RelA TAD in promoting binding to nonconsensus DNA, which sheds light on previous observations of extensive nonconsensus DNA binding by NFκB in vivo in response to strong inflammatory signals.

Cullin-independent recognition of HHARI substrates by a dynamic RBR catalytic domain

RING-between-RING (RBR) E3 ligases mediate ubiquitin transfer through an obligate E3-ubiquitin thioester intermediate prior to substrate ubiquitination. Although RBRs share a conserved catalytic module, substrate recruitment mechanisms remain enigmatic, and the relevant domains have yet to be identified for any member of the class. Here we characterize the interaction between the auto-inhibited RBR, HHARI (AriH1), and its target protein, 4EHP, using a combination of XL-MS, HDX-MS, NMR, and biochemical studies. The results show that (1) a di-aromatic surface on the catalytic HHARI Rcat domain forms a binding platform for substrates and (2) a phosphomimetic mutation on the auto-inhibitory Ariadne domain of HHARI promotes release and reorientation of Rcat for transthiolation and substrate modification. The findings identify a direct binding interaction between a RING-between-RING ligase and its substrate and suggest a general model for RBR substrate recognition.

Structural evolution of the ancient enzyme, dissimilatory sulfite reductase

Dissimilatory sulfite reductase is an ancient enzyme that has linked the global sulfur and carbon biogeochemical cycles since at least 3.47 Gya. While much has been learned about the phylogenetic distribution and diversity of DsrAB across environmental gradients, far less is known about the structural changes that occurred to maintain DsrAB function as the enzyme accompanied diversification of sulfate/sulfite reducing organisms (SRO) into new environments. Analyses of available crystal structures of DsrAB from Archaeoglobus fulgidus and Desulfovibrio vulgaris, representing early and late evolving lineages, respectively, show that certain features of DsrAB are structurally conserved, including active siro-heme binding motifs. Whether such structural features are conserved among DsrAB recovered from varied environments, including hot spring environments that host representatives of the earliest evolving SRO lineage (e.g., MV2-Eury), is not known. To begin to overcome these gaps in our understanding of the evolution of DsrAB, structural models from MV2.Eury were generated and evolutionary sequence co-variance analyses were conducted on a curated DsrAB database. Phylogenetically diverse DsrAB harbor many conserved functional residues including those that ligate active siro-heme(s). However, evolutionary co-variance analysis of monomeric DsrAB subunits revealed several False Positive Evolutionary Couplings (FPEC) that correspond to residues that have co-evolved despite being too spatially distant in the monomeric structure to allow for direct contact. One set of FPECs corresponds to residues that form a structural path between the two active siro-heme moieties across the interface between heterodimers, suggesting the potential for allostery or electron transfer within the enzyme complex. Other FPECs correspond to structural loops and gaps that may have been selected to stabilize enzyme function in different environments. These structural bioinformatics results suggest that DsrAB has maintained allosteric communication pathways between subunits as SRO diversified into new environments. The observations outlined here provide a framework for future biochemical and structural analyses of DsrAB to examine potential allosteric control of this enzyme.

Dynamics of protein kinases and pseudokinases by HDX-MS

Dynamics of the protein kinase fold are deeply intertwined with its structure. The past three decades of kinase biophysical studies revealed key dynamic features of the kinase domain and, more recently, how these features may endow catalytically impaired kinases-or pseudokinases-with signaling properties. Hydrogen-deuterium exchange coupled with mass spectrometry (HDX-MS) is proving to be a valuable approach for studies of kinase and pseudokinase domain dynamics. Here, we briefly discuss the methods that have provided insights into protein kinase dynamics, describe how HDX-MS is being used to answer questions in the kinase/pseudokinase field, and provide a detailed protocol for collecting an HDX-MS dataset to study the impacts of small molecule binding to a pseudokinase domain. As more small molecules are discovered that can disrupt pseudokinase conformations, HDX-MS is likely to be a powerful approach for exploring drug-induced changes in pseudokinase dynamics and structure.