Resistance of recombinant proteins to proteolysis during folding and in the folded state

Protein purification often involves the use of denaturing agents for solubilization. During refolding, following removal of the denaturants, the proteins of interest are exposed to proteases present in the expression system. Here the resistance of selected recombinant proteins to three widely used proteolytic enzymes, trypsin (EC 3.4.21.4), proteinase K (EC 3.4.21.14) and endoproteinase Glu-C (EC 3.4.21.19), was investigated during folding and in the folded state. Target proteins and protease mixtures were denatured in 8 mol dm^?3 urea and the proteins were allowed to refold by removal of the urea by dialysis. The proteolytic products were analyzed by sodium dodecyl sulfate–polyacrylamide gels and the protein digestion during folding was compared with the digestion under similar conditions in physiological buffer. Depending on the folding state of the proteins, the proteases had different effects on the substates. During folding, certain recombinant proteins were more efficiently digested by trypsin and, in particular, by proteinase K in comparison with digestion in the folded state, while other protein substrates were more resistant to proteolytic degradation in a denatured or partially denatured state than their folded counterparts. Incubation of most substrate proteins with endoproteinase Glu-C yielded kinetics of digestion that were essentially similar for both partially folded and unfolded substrates. The results reported may be useful for protection of sensitive proteins and in studies of protein folding mechanisms.     

Ribosome: The Structure–Function Relation and a New Paradigm to the Protein Folding Problem

The rate of protein synthesis is about seven and fifteen amino acids per second, in the eukaryotic and the bacterial ribosome, respectively. Hence, a few minutes is required to synthesize a polypeptide of an average length. This is much longer than the time needed for the hydrophobic collapse (folding) to take place. So a polypeptide gets enough time to form its local secondary to tertiary structures cotranslationally and put such segments in proper order while in association with the ribosome, unless something prevents its entire length from folding. As reported earlier, ribosomes from prokaryotes, eukaryotes, and mitochondria act as molds for protein folding, and each mold has a set of recognition sites for all proteins. More specifically, the mold is the peptidyl transferase center (PTC), a part of the large RNA of the large ribosomal subunit. Specific amino acids from different random coil regions in a protein interact with specific nucleotides in the PTC, which brings the entire length of the protein into the small space of the PTC mold. The mold thus helps to stabilize the entropy-driven collapsed state of the polypeptide. The process also divides the protein into small segments; each segment is connected at two ends with two nucleotides and can fold in the ribosomal environment. The segments dissociate in such a sequence that the organization proceeds hierarchically from the core of the globular protein radially towards the outer surface. Then the protein dissociates from the ribosome in a “folding competent state” which does the final fine tuning in folding outside the ribosome. While the ribosomal contact and release are over in 1–2 minutes in vitro, the fine tuning takes about 5–10 minutes. Release from the ribosome needs no added energy factor from outside, like ATP.     

Is Protein Folding a Thermodynamically Unfavorable,Active, Energy-Dependent Process?

The prevailing current view of protein folding is the thermodynamic hypothesis, under which the native folded conformation of a protein corresponds to the global minimum of Gibbs free energy G. We question this concept and show that the empirical evidence behind the thermodynamic hypothesis of folding is far from strong. Furthermore, physical theory-based approaches to the prediction of protein folds and their folding pathways so far have invariably failed except for some very small proteins, despite decades of intensive theory development and the enormous increase of computer power. The recent spectacular successes in protein structure prediction owe to evolutionary modeling of amino acid sequence substitutions enhanced by deep learning methods, but even these breakthroughs provide no information on the protein folding mechanisms and pathways. We discuss an alternative view of protein folding, under which the native state of most proteins does not occupy the global free energy minimum, but rather, a local minimum on a fluctuating free energy landscape. We further argue that ΔG of folding is likely to be positive for the majority of proteins, which therefore fold into their native conformations only through interactions with the energy-dependent molecular machinery of living cells, in particular, the translation system and chaperones. Accordingly, protein folding should be modeled as it occurs in vivo, that is, as a non-equilibrium, active, energy-dependent process.     

Nanosecond Dynamics of Calmodulin and Ribosome‐Bound Nascent Chains Studied by Time‐Resolved Fluorescence Anisotropy

We report a time‐resolved fluorescence anisotropy study of ribosome‐bound nascent chains (RNCs) of calmodulin (CaM), a prototypical member of the EF‐hand family of calcium‐sensing proteins. As shown in numerous studies, in vitro protein refolding can differ substantially from biosynthetic protein folding, which takes place cotranslationally and depends on the rate of polypeptide chain elongation. A challenge in this respect is to characterize the adopted conformations of nascent chains before their release from the ribosome. CaM RNCs (full‐length, half‐length, and first EF‐hand only) were synthesized in vitro. All constructs contained a tetracysteine motif site‐specifically incorporated in the first N‐terminal helix; this motif is known to react with FlAsH, a biarsenic fluorescein derivative. As the dye is rotationally locked to this helix, we characterized the structural properties and folding states of polypeptide chains tethered to ribosomes and compared these with released chains. Importantly, we observed decelerated tumbling motions of ribosome‐tethered and partially folded nascent chains, compared to released chains. This indicates a pronounced interaction between nascent chains and the ribosome surface, and might reflect chaperone activity of the ribosome.     

The Folding of Knotted Proteins: Distinguishing the Distinct Behavior of Shallow and Deep Knots

The mechanism of deep knot formation in proteins has been debated for the past two decades, but definitive answers are still lacking. In this review, we first describe knotted proteins from the perspective of shallow and deep knots, taking into account recent experimental and theoretical results. We focus on the folding mechanism, where this difference is most profound. We explain in more details the cotranslational knotting pathway. Then, we additionally show that proteins with extremely deep knots have a distinct mechanism of knotting from proteins with shallow knots. The approach based on treating shallow and deep knots as separate classes of molecules allow to classify them better and introduce a new paradigm of thinking about knotted proteins as such. This may in turn help to avoid ambiguities in further research.     

Folding of Right- and Left-Handed Three-Helix Proteins

We are the first to investigate the relationship between protein handedness and the rate of protein folding. Our findings demonstrate that small three-helix, left-handed proteins are less densely packed and should result in faster folding than that of right-handed, three-helix proteins. At the same time, right-handed, three-helix proteins have higher mechanical stability than the left-handed proteins. Moreover, from our analysis we have revealed that bacterial three-helix proteins have some advantages in packing over eukaryotic right-handed, three-helix proteins, which should result in faster folding.     

Folding core predictions from network models of proteins

Two different computational methods are employed to predict protein folding nuclei from native state structures, one based on an elastic network (EN) model and the other on a constraint network model of freely rotating rods. Three sets of folding cores are predicted with these models, and their correlation against the slow exchange folding cores identified by native state hydrogen-deuterium exchange (HX) experiments is used to test each method. These three folding core predictions rely on differences in the underlying models and relative importance of global or local motions for protein unfolding/folding reactions. For non-specific residue interactions, we use the Gaussian Network Model (GNM) to identify folding cores in the limits of two classes of motions, shortly referred to as global and local. The global mode minima from GNM represent the residues with the greatest potential for coordinating collective motions and are explored as potential folding nuclei. Additionally, the fast mode peaks that have previously been labeled as the kinetically hot residues are identified as a second folding core set dependent on local interactions. Finally, a third folding core set is defined by the most stable residues in a simulated thermal denaturation procedure of the FIRST software. This method uses an all-atomic analysis of the rigidity and flexibility of protein structures, which includes specific hydrophobic, polar and charged interactions. Comparison of the three folding core sets to HX data indicate that the fast mode peak residues determined by the GNM and the rigid folding cores of FIRST provide statistically significant enhancements over random correlation. The role of specific interactions in protein folding is also investigated by contrasting the differences between these two network-based computational methods.     

Refolding of proteins in a CSTR

A model to predict refolding of proteins in a continuous stirred tank reactor (CSTR) was developed and compared to a batch refolding process with simple dilution of the protein in a stirred tank reactor. For experimental verification of the model a continuous refolding of a model protein (α-lactalbumin) was performed in a CSTR. The refolding process of denatured and fully reduced α-lactalbumin could be accurately predicted by a set of differential equations assuming a first order reaction rate for folding and a second order reaction rate for aggregation. The system composed of a CSTR with an additional diafiltration circuit for removal of denaturing agents from the feed solution and to maintain constant refolding conditions. Based on the folding kinetic the dynamic behavior of such a continuous refolding reactor was simulated under different operating conditions. It was shown that the refolding efficiency was higher compared to batch dilution under certain conditions, namely high residence times. The yield of refolded protein could further increased by recycling the outlet stream containing unfolded protein to the reactor entrance.     

Effect of secondary structure on the conformations and folding behaviors of protein-like chains

We present a new model considering the effects of secondary structure on the conformations and folding process of protein-like chains in three-dimensional simple cubic lattice in this paper. The properties such as chain dimensions, shape, average contacts and chain average energy with different helical energy of a helix (ε_hel=0, −0.75, −1.5, and −3 in the unit of kT) are discussed here. Unlike conventional polymers, protein-like chains are much compact. We also find that the ability to form helix of residue is different under the condition of different helical energy of a helix. The energy distribution for protein-like chains with different length and the conformation changes in the process of folding of proteins are discussed. Comparisons with real protein chains are also made.     

The Folding of de Novo Designed Protein DS119 via Molecular Dynamics Simulations

As they are not subjected to natural selection process, de novo designed proteins usually fold in a manner different from natural proteins. Recently, a de novo designed mini-protein DS119, with a βαβ motif and 36 amino acids, has folded unusually slowly in experiments, and transient dimers have been detected in the folding process. Here, by means of all-atom replica exchange molecular dynamics (REMD) simulations, several comparably stable intermediate states were observed on the folding free-energy landscape of DS119. Conventional molecular dynamics (CMD) simulations showed that when two unfolded DS119 proteins bound together, most binding sites of dimeric aggregates were located at the N-terminal segment, especially residues 5–10, which were supposed to form β-sheet with its own C-terminal segment. Furthermore, a large percentage of individual proteins in the dimeric aggregates adopted conformations similar to those in the intermediate states observed in REMD simulations. These results indicate that, during the folding process, DS119 can easily become trapped in intermediate states. Then, with diffusion, a transient dimer would be formed and stabilized with the binding interface located at N-terminals. This means that it could not quickly fold to the native structure. The complicated folding manner of DS119 implies the important influence of natural selection on protein-folding kinetics, and more improvement should be achieved in rational protein design.     

Functional cell-surface display of a lipase-specific chaperone

Lipases are important enzymes in biotechnology. Extracellular bacterial lipases from Pseudomonads and related species require the assistance of specific chaperones, designated "Lif" proteins (lipase specific foldases). Lifs, a unique family of steric chaperones, are anchored to the periplasmic side of the inner membrane where they convert lipases into their active conformation. We have previously shown that the autotransporter protein EstA from P. aeruginosa can be used to direct a variety of proteins to the cell surface of Escherichia coli. Here we demonstrate for the first time the functional cell-surface display of the Lif chaperone and FACS (fluorescence-activated cell sorting)-based analysis of bacterial cells that carried foldase-lipase complexes. The model Lif protein, LipH from P. aeruginosa, was displayed at the surface of E. coli cells. Surface exposed LipH was functional and efficiently refolded chemically denatured lipase. The foldase autodisplay system reported here can be used for a variety of applications including the ultrahigh-throughput screening of large libraries of foldase variants generated by directed evolution.     

A Hamiltonian Replica Exchange Molecular Dynamics (MD) Method for the Study of Folding,Based on the Analysis of the Stabilization Determinants of Proteins

Herein, we present a novel Hamiltonian replica exchange protocol for classical molecular dynamics simulations of protein folding/unfolding. The scheme starts from the analysis of the energy-networks responsible for the stabilization of the folded conformation, by means of the energy-decomposition approach. In this framework, the compact energetic map of the native state is generated by a preliminary short molecular dynamics (MD) simulation of the protein in explicit solvent. This map is simplified by means of an eigenvalue decomposition. The highest components of the eigenvector associated with the lowest eigenvalue indicate which sites, named “hot spots”, are likely to be responsible for the stability and correct folding of the protein. In the Hamiltonian replica exchange protocol, we use modified force-field parameters to treat the interparticle non-bonded potentials of the hot spots within the protein and between protein and solvent atoms, leaving unperturbed those relative to all other residues, as well as solvent-solvent interactions. We show that it is possible to reversibly simulate the folding/unfolding behavior of two test proteins, namely Villin HeadPiece HP35 (35 residues) and Protein A (62 residues), using a limited number of replicas. We next discuss possible implications for the study of folding mechanisms via all atom simulations.     

Confirmation of the predicted source of a slow folding reaction: proline 8 of bovine pancreatic trypsin inhibitor

A major question in protein structural analysis concerns theapplicability of results from model systems to other proteins.Theoretical approaches seem the best manner of transferringinformation from one system to another, but their accuracy inthe model systems must first be tested with results from experiment.Since bovine pancreatic trypsin inhibitor (BPTI) is a modelsystem for the evaluation of energy minimization and moleculardynamics routines, we can use folding and stability measurementsto examine the reliability of these methods. All two-disulfidemutants of BPTI investigated thus far have two very slow foldingreactions which have characteristics of proline isomerization.These reactions may occur because the non-native cis form oftwo of the four prolines in BPTI significantly destabilizesthe folded state of the protein. Previous energy minimizationstudies of wild-type BPTI suggested that the cis form of Pro8was the most destabilizing of the four prolines [Levitt,M. (1981)J. Mol. Biol., 145, 251–263]. In this paper, we show thatmutation of Pro8 - Gln in the two-disulfide bond mutant Val30–Ala51results in a loss of the slowest folding reaction, consistentwith Levitt's prediction.     

Impact of the C‐terminal Disulfide Bond on the Folding and Stability of Onconase

The two homologous proteins ribonuclease A and onconase fold through conserved initial contacts but differ significantly in their thermodynamic stability. A disulfide bond is located in the folding initiation site of onconase (the C‐terminal part of the protein molecule) that is missing in ribonuclease A, whereas the other three disulfide bonds of onconase are conserved in ribonuclease A. Consequently, the deletion of this C‐terminal disulfide bond (C87–C104) allows the impact of the contacts in this region on the folding of onconase to be studied. We found the C87A/C104A‐onconase variant to be less active and less stable than the wild‐type protein, whereas the tertiary structure, which was determined by both X‐ray crystallography and NMR spectroscopy, was only marginally affected. The folding kinetics of the variant, however, were found to be changed considerably in comparison to wild‐type onconase. Proton exchange experiments in combination with two‐dimensional NMR spectroscopy revealed differences in the native‐state dynamics of the two proteins in the folding initiation site, which are held responsible for the changed folding mechanism. Likewise, the molecular dynamics simulation of the unfolding reaction indicated disparities for both proteins. Our results show that the high stability of onconase is based on the efficient stabilization of the folding initiation site by the C‐terminal disulfide bond. The formation of the on‐pathway intermediate, which is detectable during the folding of the wild‐type protein and promotes the fast and efficient refolding reaction, requires the presence of this covalent bond.     

The Finite Size Effects and Two-State Paradigm of Protein Folding

The coil to globule transition of the polypeptide chain is the physical phenomenon behind the folding of globular proteins. Globular proteins with a single domain usually consist of about 30 to 100 amino acid residues, and this finite size extends the transition interval of the coil-globule phase transition. Based on the pedantic derivation of the two-state model, we introduce the number of amino acid residues of a polypeptide chain as a parameter in the expressions for two cooperativity measures and reveal their physical significance. We conclude that the k2 measure, defined as the ratio of van ’t Hoff and calorimetric enthalpy is related to the degeneracy of the denatured state and describes the number of cooperative units involved in the transition; additionally, it is found that the widely discussed k2=1 is just the necessary condition to classify the protein as the two-state folder. We also find that Ωc, a quantity not limited from above and growing with system size, is simply proportional to the square of the transition interval. This fact allows us to perform the classical size scaling analysis of the coil-globule phase transition. Moreover, these two measures are shown to describe different characteristics of protein folding.     

Parameter optimization for the Gaussian model of protein folding

Computational models of protein folding and ligand docking are large and complex. Few systematic methods have yet been developed to optimize the parameters in such models. We describe here an iterative parameter optimization strategy that is based on minimizing a structural error measure by descent in parameter space. At the start, we know the ‘correct’ native structure that we want the model to produce, and an initial set of parameters representing the relative strengths of interactions between the amino acids. The parameters are changed systematically until the model native structure converges as closely as possible to the correct native structure. As a test, we apply this parameter optimization method to the recently developed Gaussian model of protein folding: each amino acid is represented as a bead and all bonds, covalent and noncovalent, are represented by Hooke's law springs. We show that even though the Gaussian model has continuous degrees of freedom, parameters can be chosen to cause its ground state to be identical to that of Go-type lattice models, for which the global ground states are known. Parameters for a more realistic protein model can also be obtained to produce structures close to the real native structures in the protein database.