Affinity Maturated Transferrin Receptor Apical Domain Blocks Machupo Virus Glycoprotein Binding.

Transferrin receptor 1 (TfR) delivers iron across cellular membranes by shuttling the ion carrier protein transferrin. This ability to deliver large protein ligands inside cells is taken advantage of by pathogens to infiltrate human cells. Notably, the receptor's outermost ectodomain, the apical domain, is used as a point of attachment for several viruses including hemorrhagic arenaviruses. To better understand interactions with the receptor it would be advantageous to probe sequence determinants in the apical domain with viral spike proteins. Here, we carried out affinity maturation of our computationally designed apical domain from human TfR to identify underlying driving forces that lead to better binding. The improved variants were confirmed by in vitro surface plasmon resonance measurements with dissociation constants obtained in the lower nanomolar range. It was found that the strong binding affinities for the optimized variants matched the strength of interactions with the native receptor. The structure of the best variant was determined experimentally indicating that the conformational change in the hairpin binding motif at the protein-protein interface plays a crucial role. The experimental methodology can be straightforwardly applied to other arenavirus or pathogens that use the apical domain. It can further be useful to probe host-virus compatibility or therapeutic strategies based on the transferrin receptor decoys.

Evaluation of residue variability in a conformation-specific context and during evolutionary sequence reconstruction narrows drug resistance selection in Abl1 tyrosine kinase.

Diseases with readily available therapies may eventually prevail against the specific treatment by the acquisition of resistance. The constitutively active Abl1 tyrosine kinase known to cause chronic myeloid leukemia is an example, where patients may experience relapse after small inhibitor drug treatment. Mutations in the Abl1 tyrosine kinase domain (Abl1-KD) are a critical source of resistance and their emergence depends on the conformational states that have been observed experimentally: the inactive state, the active state, and the intermediate inactive state that resembles Src kinase. Understanding how resistant positions and amino acid identities are determined by selection pressure during drug treatment is necessary to improve future drug development or treatment decisions. We carry out in silico site-saturation mutagenesis over the Abl1-KD structure in a conformational context to evaluate the in situ and conformational stability energy upon mutation. Out of the 11 studied resistant positions, we determined that 7 of the resistant mutations favored the active conformation of Abl1-KD with respect to the inactive state. When, instead, the sequence optimization was modeled simultaneously at resistant positions, we recovered five known resistant mutations in the active conformation. These results suggested that the Abl1 resistance mechanism targeted substitutions that favored the active conformation. Further sequence variability, explored by ancestral reconstruction in Abl1-KD, showed that neutral genetic drift, with respect to amino acid variability, was specifically diminished in the resistant positions. Since resistant mutations are susceptible to chance with a certain probability of fixation, combining methodologies outlined here may narrow and limit the available sequence space for resistance to emerge, resulting in more robust therapeutic treatments over time.

Motif-driven protein binder design towards transferrin receptor helical domain.

Human transferrin receptor 1 (TfR) is necessary for the delivery of the iron carrier protein transferrin into cells and can be utilized for targeted delivery across cellular membranes. Binding of transferrin to the receptor is regulated by hereditary hemochromatosis protein (HFE), an iron regulatory protein that partly shares a binding site with transferrin on TfR. Here, we derived essential binding interactions from HFE and computationally grafted these into a library of small protein scaffolds. One of the designed proteins, TB08, was further optimized computationally and experimentally to identify variants with improved binding to TfR. The optimized variant, TB08 S3.1, expressed well in the E. coli expression system and had an affinity to TfR in the low micromolar range, Kd ≈ 1 µm, as determined by surface plasmon resonance. A binding competition assay with transferrin further confirmed the interaction of the evolved variant to TfR at the shared binding surface. Additionally, the GFP-tagged evolved variant of TB08 demonstrated cellular internalization as determined by fluorescent and confocal microscopy in HeLa cells. The designed protein is small, allows for robust cargo tagging, and interacts specifically with TfR, thus making it a valuable tool for the characterization of TfR-mediated cellular transport mechanisms and for the assessment of engineering strategies for cargo delivery across cell membranes.

Prediction of Protein-Protein Binding Interactions in Dimeric Coiled Coils by Information Contained in Folding Energy Landscapes.

Coiled coils represent the simplest form of a complex formed between two interacting protein partners. Their extensive study has led to the development of various methods aimed towards the investigation and design of complex forming interactions. Despite the progress that has been made to predict the binding affinities for protein complexes, and specifically those tailored towards coiled coils, many challenges still remain. In this work, we explore whether the information contained in dimeric coiled coil folding energy landscapes can be used to predict binding interactions. Using the published SYNZIP dataset, we start from the amino acid sequence, to simultaneously fold and dock approximately 1000 coiled coil dimers. Assessment of the folding energy landscapes showed that a model based on the calculated number of clusters for the lowest energy structures displayed a signal that correlates with the experimentally determined protein interactions. Although the revealed correlation is weak, we show that such correlation exists; however, more work remains to establish whether further improvements can be made to the presented model.

Tuning the binding interface between Machupo virus glycoprotein and human transferrin receptor.

Machupo virus, known to cause hemorrhagic fevers, enters human cells via binding with its envelope glycoprotein to transferrin receptor 1 (TfR). Similarly, the receptor interactions have been explored in biotechnological applications as a molecular system to ferry therapeutics across the cellular membranes and through the impenetrable blood-brain barrier that effectively blocks any such delivery into the brain. Study of the experimental structure of Machupo virus glycoprotein 1 (MGP1) in complex with TfR and glycoprotein sequence homology has identified some residues at the interface that influence binding. There are, however, no studies that have attempted to optimize the binding potential between MGP1 and TfR. In pursuits for finding therapeutic solutions for the New World arenaviruses, and to gain a greater understanding of MGP1 interactions with TfR, it is crucial to understand the structure-sequence relationship driving the interface formation. By displaying MGP1 on yeast surface we have examined the contributions of individual residues to the binding of solubilized ectodomain of TfR. We identified MGP1 binding hot spot residues, assessed the importance of posttranslational N-glycan modifications, and used a selection with random mutagenesis for affinity maturation. We show that the optimized MGP1 variants can bind more strongly to TfR than the native MGP1, and there is an MGP1 sequence that retains binding in the absence of glycosylation, but with the addition of further amino acid substitutions. The engineered variants can be used to probe cellular internalization or the blood-brain barrier crossing to achieve greater understanding of TfR mediated internalization.

Computational backbone design enables soluble engineering of transferrin receptor apical domain.

Supply of iron into human cells is achieved by iron carrier protein transferrin and its receptor that upon complex formation get internalized by endocytosis. Similarly, the iron needs to be delivered into the brain, and necessitates the transport across the blood-brain barrier. While there are still unanswered questions about these mechanisms, extensive efforts have been made to use the system for delivery of therapeutics into biological compartments. The dimeric form of the receptor, where each subunit consists of three domains, further complicates the detailed investigation of molecular determinants responsible for guiding the receptor interactions with other proteins. Especially the apical domain's biological function has been elusive. To further the study of transferrin receptor, we have computationally decoupled the apical domain for soluble expression, and validated the design strategy by structure determination. Besides presenting a methodology for solubilizing domains, the results will allow for study of apical domain's function.

Deciphering the molecular mechanism of FLT3 resistance mutations.

FMS-like tyrosine kinase 3 (FLT3) has been found to be mutated in ~ 30% of acute myeloid leukaemia patients. Small-molecule inhibitors targeting FLT3 that are currently approved or still undergoing clinical trials are subject to drug resistance due to FLT3 mutations. How these mutations lead to drug resistance is hitherto poorly understood. Herein, we studied the molecular mechanism of the drug resistance mutations D835N, Y842S and M664I, which confer resistance against the most advanced inhibitors, quizartinib and PLX3397 (pexidartinib), using enzyme kinetics and computer simulations. In vitro kinase assays were performed to measure the comparative catalytic activity of the native protein and the mutants, using a bacterial expression system developed to this aim. Our results reveal that the differential drug sensitivity profiles can be rationalised by the dynamics of the protein-drug interactions and perturbation of the intraprotein contacts upon mutations. Drug binding induced a single conformation in the native protein, whereas multiple conformations were observed otherwise (in the mutants or in the absence of drugs). The end-point kinetics measurements indicated that the three resistant mutants conferred catalytic activity that is at least as high as that of the reference without such mutations. Overall, our calculations and measurements suggest that the structural dynamics of the drug-resistant mutants that affect the active state and the increased conformational freedom of the remaining inactive drug-bound population are the two major factors that contribute to drug resistance in FLT3 harbouring cancer cells. Our results explain the mechanism of drug resistance mutations and can aid to the design of more effective tyrosine kinase inhibitors.

The catalytic activity of Abl1 single and compound mutations: Implications for the mechanism of drug resistance mutations in chronic myeloid leukaemia.

BACKGROUND: Abl1 is a protein tyrosine kinase whose aberrant activation due to mutations is the culprit of several cancers, most notably chronic myeloid leukaemia. Several Abl1 inhibitors are used as anti-cancer drugs. Unfortunately, drug resistance limits their effectiveness. The main cause for drug resistance is mutations in the kinase domain (KD) of Abl1 that evolve in patients. The T315I mutation confers resistance against all clinically-available inhibitors except ponatinib. Resistance to ponatinib can develop by compound (double) mutations. METHODS: Kinetic measurements of the KD of Abl1 and its mutants were carried out to examine their catalytic activity. Specifically, mutants that lead to drug resistance against ponatinib were considered. Molecular dynamics simulations and multiple sequence analysis were used for explanation of the experimental findings. RESULTS: The catalytic efficiency of the T315I pan-resistance mutant is more than two times lower than that of the native KD. All ponatinib resistant mutations restore the catalytic efficiency of the enzyme. Two of them (G250E/T315I and Y253H/E255V) have a catalytic efficiency that is more than five times that of the native KD. CONCLUSIONS: The measurements and analysis suggest that resistance is at least partially due to the development of a highly efficient kinase through subsequent mutations. The simulations highlight modifications in two structurally important regions of Abl1, the activation and phosphate binding loops, upon mutations. GENERAL SIGNIFICANCE: Experimental and computational methods were used together to explain how mutations in the kinase domain of Abl1 lead to resistance against the most advanced drug currently in use to treat chronic myeloid leukaemia.

Computational assessment of folding energy landscapes in heterodimeric coiled coils.

The coiled coil structural motif consists of alpha helices supercoiling around each other to form staggered knobs-into-holes packing. Such structures are deceptively simple, especially as they often can be described with parametric equations, but are known to exist in various conformations. Even the simplest systems, consisting of 2 monomers, can assemble into a wide range of states. They can form canonical as well as noncanonical coiled coils, be parallel or antiparallel, where helices associate with different degrees of shift, tilt, and rotation. Here, we investigate the energy landscape of heterodimeric coiled coils by carrying out de novo folding simulations starting from amino acid sequence. We folded a diverse set of 22 heterodimers and demonstrate that the approach is capable of identifying the atomic details in the experimental structure in the majority of cases. Our methodology also enables exploration of alternative states that can be accessible in solution beyond the experimentally determined structure. For many systems, we observe folding energy landscapes with multiple energy minima and several isoenergetic states. By comparing coiled coils from single domains and those extracted from larger proteins, we find that standalone coiled coils have deeper energy wells at the experimentally determined conformation. By folding the competing homodimeric states in addition to the heterodimers, we observe that the structural specificity towards the heteromeric state is often small. Taken together, our results demonstrate that de novo folding simulations can be a powerful tool to characterize structural specificity of coiled coils when coupled to assessment of energy landscapes.

Enantioselective enzymes by computational design and in silico screening.

Computational enzyme design holds great promise for providing new biocatalysts for synthetic chemistry. A strategy to design small mutant libraries of complementary enantioselective epoxide hydrolase variants for the production of highly enantioenriched (S,S)-diols and (R,R)-diols is developed. Key features of this strategy (CASCO, catalytic selectivity by computational design) are the design of mutations that favor binding of the substrate in a predefined orientation, the introduction of steric hindrance to prevent unwanted substrate binding modes, and ranking of designs by high-throughput molecular dynamics simulations. Using this strategy we obtained highly stereoselective mutants of limonene epoxide hydrolase after experimental screening of only 37 variants. The results indicate that computational methods can replace a substantial amount of laboratory work when developing enantioselective enzymes.

De novo-designed enzymes as small-molecule-regulated fluorescence imaging tags and fluorescent reporters.

Enzyme-based tags attached to a protein-of-interest (POI) that react with a small molecule, rendering the conjugate fluorescent, are very useful for studying the POI in living cells. These tags are typically based on endogenous enzymes, so protein engineering is required to ensure that the small-molecule probe does not react with the endogenous enzyme in the cell of interest. Here we demonstrate that de novo-designed enzymes can be used as tags to attach to POIs. The inherent bioorthogonality of the de novo-designed enzyme-small-molecule probe reaction circumvents the need for protein engineering, since these enzyme activities are not present in living organisms. Herein, we transform a family of de novo-designed retroaldolases into variable-molecular-weight tags exhibiting fluorescence imaging, reporter, and electrophoresis applications that are regulated by tailored, reactive small-molecule fluorophores.

Small molecule probes to quantify the functional fraction of a specific protein in a cell with minimal folding equilibrium shifts.

Although much is known about protein folding in buffers, it remains unclear how the cellular protein homeostasis network functions as a system to partition client proteins between folded and functional, soluble and misfolded, and aggregated conformations. Herein, we develop small molecule folding probes that specifically react with the folded and functional fraction of the protein of interest, enabling fluorescence-based quantification of this fraction in cell lysate at a time point of interest. Importantly, these probes minimally perturb a protein's folding equilibria within cells during and after cell lysis, because sufficient cellular chaperone/chaperonin holdase activity is created by rapid ATP depletion during cell lysis. The folding probe strategy and the faithful quantification of a particular protein's functional fraction are exemplified with retroaldolase, a de novo designed enzyme, and transthyretin, a nonenzyme protein. Our findings challenge the often invoked assumption that the soluble fraction of a client protein is fully folded in the cell. Moreover, our results reveal that the partitioning of destabilized retroaldolase and transthyretin mutants between the aforementioned conformational states is strongly influenced by cytosolic proteostasis network perturbations. Overall, our results suggest that applying a chemical folding probe strategy to other client proteins offers opportunities to reveal how the proteostasis network functions as a system to regulate the folding and function of individual client proteins in vivo.

Automating human intuition for protein design.

In the design of new enzymes and binding proteins, human intuition is often used to modify computationally designed amino acid sequences prior to experimental characterization. The manual sequence changes involve both reversions of amino acid mutations back to the identity present in the parent scaffold and the introduction of residues making additional interactions with the binding partner or backing up first shell interactions. Automation of this manual sequence refinement process would allow more systematic evaluation and considerably reduce the amount of human designer effort involved. Here we introduce a benchmark for evaluating the ability of automated methods to recapitulate the sequence changes made to computer-generated models by human designers, and use it to assess alternative computational methods. We find the best performance for a greedy one-position-at-a-time optimization protocol that utilizes metrics (such as shape complementarity) and local refinement methods too computationally expensive for global Monte Carlo (MC) sequence optimization. This protocol should be broadly useful for improving the stability and function of designed binding proteins.

Exploration of alternate catalytic mechanisms and optimization strategies for retroaldolase design.

Designed retroaldolases have utilized a nucleophilic lysine to promote carbon-carbon bond cleavage of ß-hydroxy-ketones via a covalent Schiff base intermediate. Previous computational designs have incorporated a water molecule to facilitate formation and breakdown of the carbinolamine intermediate to give the Schiff base and to function as a general acid/base. Here we investigate an alternative active-site design in which the catalytic water molecule was replaced by the side chain of a glutamic acid. Five out of seven designs expressed solubly and exhibited catalytic efficiencies similar to previously designed retroaldolases for the conversion of 4-hydroxy-4-(6-methoxy-2-naphthyl)-2-butanone to 6-methoxy-2-naphthaldehyde and acetone. After one round of site-directed saturation mutagenesis, improved variants of the two best designs, RA114 and RA117, exhibited among the highest kcat (>10(-3)s(-1)) and kcat/KM (11-25M(-1)s(-1)) values observed for retroaldolase designs prior to comprehensive directed evolution. In both cases, the >10(5)-fold rate accelerations that were achieved are within 1-3 orders of magnitude of the rate enhancements reported for the best catalysts for related reactions, including catalytic antibodies (kcat/kuncat=10(6) to 10(8)) and an extensively evolved computational design (kcat/kuncat>10(7)). The catalytic sites, revealed by X-ray structures of optimized versions of the two active designs, are in close agreement with the design models except for the catalytic lysine in RA114. We further improved the variants by computational remodeling of the loops and yeast display selection for reactivity of the catalytic lysine with a diketone probe, obtaining an additional order of magnitude enhancement in activity with both approaches.

Computational design of enone-binding proteins with catalytic activity for the Morita-Baylis-Hillman reaction.

The Morita-Baylis-Hillman reaction forms a carbon-carbon bond between the α-carbon of a conjugated carbonyl compound and a carbon electrophile. The reaction mechanism involves Michael addition of a nucleophile catalyst at the carbonyl ß-carbon, followed by bond formation with the electrophile and catalyst disassociation to release the product. We used Rosetta to design 48 proteins containing active sites predicted to carry out this mechanism, of which two show catalytic activity by mass spectrometry (MS). Substrate labeling measured by MS and site-directed mutagenesis experiments show that the designed active-site residues are responsible for activity, although rate acceleration over background is modest. To characterize the designed proteins, we developed a fluorescence-based screen for intermediate formation in cell lysates, carried out microsecond molecular dynamics simulations, and solved X-ray crystal structures. These data indicate a partially formed active site and suggest several clear avenues for designing more active catalysts.

De novo enzyme design using Rosetta3.

The Rosetta de novo enzyme design protocol has been used to design enzyme catalysts for a variety of chemical reactions, and in principle can be applied to any arbitrary chemical reaction of interest. The process has four stages: 1) choice of a catalytic mechanism and corresponding minimal model active site, 2) identification of sites in a set of scaffold proteins where this minimal active site can be realized, 3) optimization of the identities of the surrounding residues for stabilizing interactions with the transition state and primary catalytic residues, and 4) evaluation and ranking the resulting designed sequences. Stages two through four of this process can be carried out with the Rosetta package, while stage one needs to be done externally. Here, we demonstrate how to carry out the Rosetta enzyme design protocol from start to end in detail using for illustration the triosephosphate isomerase reaction.

Cold adaptation of enzyme reaction rates.

A major issue for organisms living at extreme temperatures is to preserve both stability and activity of their enzymes. Cold-adapted enzymes generally have a reduced thermal stability, to counteract freezing, and show a lower enthalpy and a more negative entropy of activation compared to mesophilic and thermophilic homologues. Such a balance of thermodynamic activation parameters can make the reaction rate decrease more linearly, rather than exponentially, as the temperature is lowered, but the structural basis for rate optimization toward low working temperatures remains unclear. In order to computationally address this problem, it is clear that reaction simulations rather than standard molecular dynamics calculations are needed. We have thus carried out extensive computer simulations of the keto-enol(ate) isomerization steps in differently adapted citrate synthases to explore the structure-function relationships behind catalytic rate adaptation to different temperatures. The calculations reproduce the absolute rates of the psychrophilic and mesophilic enzymes at 300 K, as well as the lower enthalpy and more negative entropy of activation of the cold-adapted enzyme, where the latter simulation result is obtained from high-precision Arrhenius plots. The overall catalytic effect originates from electrostatic stabilization of the transition state and enolate and the reduction of reorganization free energy. The simulations, however, show psychrophilic, mesophilic, and hyperthermophilic citrate synthases to have increasingly stronger electrostatic stabilization of the transition state, while the energetic penalty in terms of internal protein interactions follows the reverse order with the cold-adapted enzyme having the most favorable energy term. The lower activation enthalpy and more negative activation entropy observed for cold-adapted enzymes are found to be associated with a decreased protein stiffness. The origin of this effect is, however, not localized to the active site but to other regions of the protein structure.

Catalysis and linear free energy relationships in aspartic proteases.

Aspartic proteases are receiving considerable attention as potential drug targets in several serious diseases, such as AIDS, malaria, and Alzheimer's disease. These enzymes cleave polypeptide chains, often between specific amino acid residues, but despite the common reaction mechanism, they exhibit large structural differences. Here, the catalytic mechanism of aspartic proteases plasmepsin II, cathepsin D, and HIV-1 protease is examined by computer simulations utilizing the empirical valence bond approach in combination with molecular dynamics and free energy perturbation calculations. Free energy profiles are established for four different substrates, each six amino acids long and containing hydrophobic side chains in the P1 and P1' positions. Our simulations reproduce the catalytic effect of these enzymes, which accelerate the reaction rate by a factor of approximately 10(10) compared to that of the corresponding uncatalyzed reaction in water. The calculations elucidate the origin of the catalytic effect and allow a rationalization of the fact that, despite large structural differences between plasmepsin II/cathepsin D and HIV-1 protease, the magnitude of their rate enhancement is very similar. Amino acid residues surrounding the active site together with structurally conserved water molecules are found to play an important role in catalysis, mainly through dipolar (electrostatic) stabilization. A linear free energy relationship for the reactions in the different enzymes is established that also demonstrates the reduced reorganization energy in the enzymes compared to that in the uncatalyzed water reaction.

Computational prediction of structure, substrate binding mode, mechanism, and rate for a malaria protease with a novel type of active site.

The histo-aspartic protease (HAP) from the malaria parasite P. falciparum is one of several new promising targets for drug intervention. The enzyme possesses a novel type of active site, but its 3D structure and mechanism of action are still unknown. Here we use a combination of homology modeling, automated docking searches, and molecular dynamics/reaction free energy profile simulations to predict the enzyme structure, conformation of bound substrate, catalytic mechanism, and rate of the peptide cleavage reaction. We find that the computational tools are sufficiently reliable both for identifying substrate binding modes and for distinguishing between different possible reaction mechanisms. It is found that the favored pathway only involves direct participation by the catalytic aspartate, with the neighboring histidine providing critical stabilization (by a factor of approximately 10000) along the reaction. The calculated catalytic rate constant of about 0.1 s(-1) for a hexapeptide substrate derived from the alpha chain of human hemoglobin is in excellent agreement with experimental kinetic data for a similar peptide fragment.

Potent inhibitors of the Plasmodium falciparum enzymes plasmepsin I and II devoid of cathepsin D inhibitory activity.

The hemoglobin-degrading aspartic proteases plasmepsin I (Plm I) and plasmepsin II (Plm II) of the malaria parasite Plasmodium falciparum have lately emerged as putative drug targets. A series of C(2)-symmetric compounds encompassing the 1,2-dihydroxyethylene scaffold and a variety of elongated P1/P1' side chains were synthesized via microwave-assisted palladium-catalyzed coupling reactions. Binding affinity calculations with the linear interaction energy method and molecular dynamics simulations reproduced the experimental binding data obtained in a Plm II assay with very good accuracy. Bioactive conformations of the elongated P1/P1' chains were predicted and agreed essentially with a recent X-ray structure. The compounds exhibited picomolar to nanomolar inhibition constants for the plasmepsins and no measurable affinity to the human enzyme cathepsin D. Some of the compounds also demonstrated significant inhibition of parasite growth in cell culture. To the best of our knowledge, these plasmepsin inhibitors represent the most selective reported to date and constitute promising lead compounds for further optimization.