Unfolding Rate Research Articles

Complete Extraction of Protein Dynamics Information in Hydrogen/Deuterium Exchange Mass Spectrometry Data.

Hydrogen/deuterium exchange (HDX) mass spectrometry (MS) has been used to study protein conformation and conformational dynamics. A continuous labeling experiment, followed by proteolytic digestion and MS analysis, generates a large amount of data, containing information on protein conformation and conformational dynamics. Lacking appropriate computational methods, information hidden inside the isotope distribution is often omitted and not extracted. In this work, a computational model is described to simulate the determined isotope pattern for each proteolytic peptide at each labeling time. Optimizing the model with experimental data yields conformational protection as well as protein unfolding/folding kinetics. With this method, complete extraction of protein dynamics information in the HDX-MS data is achieved. Information derived from the method is reliable as the model is mostly based on first-principles with very few assumptions. It is demonstrated that the protein dynamics information extracted from one or two labeling time points approaches or exceeds the information derived from an entire deuterium uptake time course by the traditional method. Application of the computational method to an IgG1 antibody under mild denaturing conditions indicates that the unfolding of each immunoglobulin domain can be explained by a simple two-state unfolding process, with different unfolding rate for each domain.

The Effect of Chain Connectivity on the Thermodynamic, Kinetic and Mechanical Properties of Azurin

Azurin, a blue-copper protein, is responsible for electron transfer in Pseudomonas aeruginosa. It has widely been studied for its stability, electronic and electron transport properties, with potential applications in biomaterials-based devices. The amino acid chain connectivity is of fundamental importance in modulating the protein stability and folding. Hence, here we studied the thermodynamic, kinetic and mechanical properties of a circular permutant of apo-azurin (Azu-CPN42), with its N-terminus at Asn42, and compared them with those of the wild-type apo-azurin (WT-Azu). While the spectroscopic and copper-binding properties of Azu-CPN42 are unaltered, we observe changes in protein stability. Equilibrium unfolding studies using GdHCl suggest a drop of ∼0.6M in Cm and ∼13 kJ/mol in free energy of unfolding for Azu-CPN42 compared to that of WT-Azu. Kinetic studies reveal increase in the spontaneous unfolding rate by two-orders of magnitude, while the spontaneous folding rate being similar. Mechanical properties probed using single-molecule atomic force microscope suggest parallel unfolding pathways via 2- and 3-states, similar to that of WT-Azu reported previously. Interestingly, the intermediate stability is lower than that of the native state of Azu-CPN42 by ∼10pN, in accord with that of WT-Azu. As anticipated, the mechanical unfolding pathways of Azu-CPN42 and WT-Azu are different, as the two proteins are pulled along different N-C termini. This resulted in different mechanical intermediates, with a broad distribution of unfolding contour lengths (∼3-25 nm) for Azu-CPN42 while those of WT-Azu are centered at 7.5 ± 1.5 nm. This huge diversity may arise due to mechanical unraveling from both C- and N- termini as opposed to only C-terminal unraveling observed in WT-Azu. Our study reveals a decreased thermodynamic and kinetic stability and altered mechanical unfolding pathway of azurin upon changing its chain connectivity through circular permutation.

How Quickly Do Proteins Fold and Unfold, and What Structural Parameters Correlate with These Values?

The correlations between the logarithm of the unfolding rate of 108 proteins and their structural parameters were calculated. We showed that there is a good correlation between the logarithm of folding and unfolding rates (0.79) and protein stability and unfolding rate (0.79). Thus, the faster the protein folds, the faster it unfolds. Folding and unfolding rates are higher for the proteins with two-state kinetics, in comparison with the proteins with multi-state kinetics. At the same time, two-state bacterial proteins folds and unfolds two orders of magnitude faster than two-state eukaryotic proteins, and multi-state bacterial proteins folds and unfolds slower than multi-state eukaryotic proteins. Despite the fact that the folding rates of thermophilic and mesophilic proteins are close, the unfolding rates of thermophilic proteins is about two orders of magnitude lower than for mesophilic proteins. The correlation between unfolding rate and stability of thermophilic proteins is high (0.90). We also found that the unfolding rate correlates with such structural parameters as: size of the protein, radius of the cross-section, logarithm of absolute contact order, and radius of gyration. This information will be useful for engineering and designing new proteins with desired properties.

Simulation of thrombosis in a stenotic microchannel: The effects of vWF-enhanced shear activation of platelets

This study was undertaken to develop a numerical/computational simulation of von Willebrand Factor (vWF) - mediated platelet shear activation and deposition in an idealized stenosis. Blood is treated as a multi-constituent mixture comprised of a linear fluid component and a porous solid component (thrombus). Chemical and biological species involved in coagulation are modeled using a system of coupled convection-reaction-diffusion (CRD) equations. This study considers the cumulative effect of shear stress (history) on platelet activation. The vWF activity is modeled as an enhancement function for the shear stress accumulation and is related to the experimentally-observed unfolding rate of vWF. A series of simulations were performed in an idealized stenosis in which the predicted platelets deposition agreed well with previous experimental observations spatially and temporally, including the reduction of platelet deposition with decreasing expansion angle. Further simulation indicated a direct relationship between vWF-mediated platelet deposition and degree of stenosis. Based on the success with these benchmark simulations, it is hoped that the model presented here may provide additional insight into vWF-mediated thrombosis and prove useful for the development of more hemo-compatible blood-wetted devices in the future.

Mechanical unfolding of spectrin reveals a super-exponential dependence of unfolding rate on force

We investigated the mechanical unfolding of single spectrin molecules over a broad range of loading rates and thus unfolding forces by combining magnetic tweezers with atomic force microscopy. We find that the mean unfolding force increases logarithmically with loading rate at low loading rates, but the increase slows at loading rates above 1pN/s. This behavior indicates an unfolding rate that increases exponentially with the applied force at low forces, as expected on the basis of one-dimensional models of protein unfolding. At higher forces, however, the increase of the unfolding rate with the force becomes faster than exponential, which may indicate anti-Hammond behavior where the structures of the folded and transition states become more different as their free energies become more similar. Such behavior is rarely observed and can be explained by either a change in the unfolding pathway or as a reflection of a multidimensional energy landscape of proteins under force.

Kinetic Stability of Long-Lived Human Lens γ-Crystallins and Their Isolated Double Greek Key Domains

The γ-crystallins of the eye lens nucleus are among the longest-lived proteins in the human body. Synthesized in utero, they must remain folded and soluble throughout adulthood to maintain lens transparency and avoid cataracts. γD- and γS-crystallin are two major monomeric crystallins of the human lens. γD-crystallin is concentrated in the oldest lens fiber cells, the lens nucleus, whereas γS-crystallin is concentrated in the younger cells of the lens cortex. The kinetic stability parameters of these two-domain proteins and their isolated domains were determined and compared. Kinetic unfolding experiments monitored by fluorescence spectroscopy in varying concentrations of guanidinium chloride were used to extrapolate unfolding rate constants and half-lives of the crystallins in the absence of the denaturant. Consistent with their long lifespans in the lens, extrapolated half-lives for the initial unfolding step were on the timescale of years. Both proteins’ isolated N-terminal domains were less kinetically stable than their respective C-terminal domains at denaturant concentrations predicted to disrupt the domain interface, but at low denaturant concentrations, the relative kinetic stabilities were reversed. Cataract-associated aggregation has been shown to proceed from partially unfolded intermediates in these proteins; their extreme kinetic stability likely evolved to protect the lens from the initiation of aggregation reactions. Our findings indicate that the domain interface is the source of significant kinetic stability. The gene duplication and fusion event that produced the modern two-domain architecture of vertebrate lens crystallins may be the origin of their high kinetic as well as thermodynamic stability.

Breakdown of the scallop theorem for an asymmetrical folding molecular motor in soft matter.

We use molecular dynamic simulations to investigate the motion of a folding molecular motor inside soft matter. Purcell's scallop theorem forbids the displacement of the motor due to time symmetrical hydrodynamic laws at low Reynolds numbers whatever the asymmetry of the folding and unfolding rates. However, the fluctuation theorems imply a violation of the time symmetry of the motor's trajectories due to the entropy generated by the motor, suggesting a breakdown of the scallop theorem at the nanoscale. To clarify this picture, we study the predicted violation of time reversibility of the motor's trajectories, using two reverse asymmetric folding mechanisms. We actually observe this violation of time reversibility of the motor's trajectories. We also observe the previously reported fluidization of the medium induced by the motor's folding, but find that this induced diffusion is not enough to explain the increase of the motor's displacement. As a result, the motor is not carried by the medium in our system but moves by its own, in violation of the scallop theorem. The observed violation of the scallop theorem opens a route to create very simple molecular motors moving in soft matter environments.

Folding/unfolding kinetics of G-quadruplexes upstream of the P1 promoter of the human BCL-2 oncogene

The G-rich Pu39 region of the P1 promoter of the oncogene BCL-2, an apoptosis regulator, can fold into multiple G-quadruplex (G4) structures. Bcl2-2345 and Bcl2-1245 are two major G4 species forming with high thermal stability and distinct topologies in the Pu39 region, but their folding/unfolding kinetics have not yet been investigated. Here, we used magnetic tweezers to measure the mechanical stability and the folding/unfolding kinetics of the Bcl2-2345 and Bcl2-1245 G4 structures. We report that the hybrid-stranded Bcl2-2345 G4 had a lower mechanical stability than the parallel-stranded Bcl2-1245 G4. We observed that the Bcl2-2345 G4 is a kinetically favored structure, whereas the Bcl2-1245 G4, with a slow unfolding rate, may function as a kinetic barrier for transcription. We also determined that in addition to the Bcl2-2345 and Bcl2-1245 G4s, other stable DNA secondary structures, such as a hybrid-stranded Bcl2-1234 G4, can also form in the Pu39 sequence. The characterization of the folding/unfolding kinetics of specific G4s reported here sheds light on the participation of G4s during gene transcription and provides information for designing G4-targeting small molecules that could modulate BCL-2 gene expression.

Selection tools for reducing generation time of geophytic herbaceous perennials

Lengthy generation times in herbaceous perennials (1 to 5-10+ years from seed to flowering) impede the rate of progress flower breeders can make during breeding and/or domestication of floricultural crops. Geophytic herbaceous perennials, i.e., those with over-wintering underground storage organs (rhizomes, bulbs, corms, tubers etc.), have the added complication of the need for meristem size for vegetative to reproductive phase change competency and increased flower bud number. Consequently, these and other factors necessitate that most geophytic floricultural crops are primarily vegetatively (asexually) propagated. The University of Minnesota Flower Breeding and Genetics program has focused>20 years to reduce the generation time in many geophytic crops, including monocots (Gladiolus, Lilium, Iris) and dicots (Chrysanthemum), using a suite of selection techniques from seed germination onwards to flowering. Our objectives are to compare the selection techniques which have accomplished these goals: time (week) of seed germination, rate of subsequent growth, contractile root formation (corm and bulb crops), leaf number, leaf unfolding rates, stem elongation, stem and inflorescence height, internode number and length, propagule formation (size, weight), visible bud date and flowering, and flower number. Germinating seedlings were tooth picked each week with a different color each week to designate germination weeks (G1 for week 1, etc.) which enabled selecting and transplanting only the earliest germinators/generation. In most crops (Gladiolus, Iris, Chrysanthemum), seed germination during weeks 2-4 after sowing (G2-G4) was highly correlated with visible bud dates and flowering, even if early generations flowered after Year 1. In Lilium, however, G1-G2 seedlings had earlier flowering than those in G3-G8. Incorporation of these techniques along with fast cycling within geophytic breeding programs can enhance the rate of progress made per generation, particularly when the annual cycle is reduced to 3-4 months instead of 12. Comparative analyses of traits within and among each crop allows for creating crop-specific strategies to reduce generation time and maximize gain from selection.

A Decoy Folding Nucleus can Modulate Protein Folding Kinetics

The ability to modulate the folding kinetics of a specific protein in a milieu can be an important tool for chemical biology and pharmacology. Kinetics and thermodynamics of folding are routinely modulated by changing solvent conditions, but such changes are not protein specific. Here we demonstrate a general principle by which specific kinetic modulation can be achieved for any protein whose folding route is known. Our strategy is to design a pre-stabilized folding nucleus of a protein. We show that the β1- β2 part (residues 1 to 17) of ubiquitin, which is the segment known to fold first (into a β-hairpin shape)[1], can make folding faster when introduced as a separate conformationally constrained peptide at excess concentrations. A weaker but still significant effect is observed when this ‘decoy nucleus’ peptide is not conformationally constrained, or is in an unfolding buffer to start with. Interestingly, the thermodynamic stability of the protein remains unchanged, as the unfolding rate also becomes faster. We simulate the system for both a weakly constrained and a strongly constrained decoy nucleus, and observe that the decoy interacts strongly with the protein only when it is conformationally constrained. Together, our results suggest that the folding of almost any protein which possesses a well-defined folding nucleus can be modulated, by introducing a nucleus-mimicking molecule with a stabilized structure. It also implies that in vivo, fragments of proteins which are naturally present as degradation by-products can in principle affect the folding of the full length proteins. [1] Atomic-level description of ubiquitin folding. Piana et al., PNAS, 2013, 110 (15), 5915-5920

Detailed Kinetics of RNA Folding Pathways and Thermodynamic Origins of Crowding by Single-Molecule FRET

Specific three dimensional folding is required for RNA function while RNA can be a highly dynamic system that actively samples different configurations under thermal fluctuations. For instance, in gene regulation of RNA riboswitch, the prompt structural transition in response to ligand is crucial for triggering transcriptional termination, making the time scale of folding and ligand response exceedingly relevant to RNA functions. In single-molecule FRET (smFRET) experiments, the conformational changes of RNA are visualized by the time-dependent energy transfer to determine folding and unfolding rates. Such kinetic information allows us to construct a model for manganese riboswitch folding and explore the detailed folding mechanisms, where Mg2+ binding is found to pre-fold the RNA, followed by ligand (Mn2+) association to further stabilize the folded structure. Moreover, the thermodynamic origins can be obtained from temperature dependent folding. The manganese riboswitch folding is found unexpectedly promoted by heat and the process is endothermic with net entropy gain, much contradictory to the conventional idea of folding into a more order state. In molecular crowding study where we try to understand the influence of fundamental excluded volume on RNA structures in highly concentrated intracellular environments, the chemically inert crowders are shown to entropically favor the more compact folded structure of RNA; a simple physical picture of osmotic pressure is employed to explain the crowder size dependence. Lastly, a hydraulic pressure generator is used to apply pressure (up to 5 kbar) to RNA sample through capillary. The free volume change (pΔV) during RNA folding may be related to structural response under crowding conditions.

G-Quadruplex-Helicase Interactions and the Impact of Small Molecules

One of the therapeutic targets that has attracted attention in targeted therapy is guanine quadruplex (GQ), which forms in guanine-rich nucleic acid sequences, such as telomeres. Studies on cancer cells show that stabilizing the GQ structures in telomeres with small molecules reduces telomerase activity and proliferation of cancer cells. Because of their role in biological processes and their potential use as therapeutic targets, GQ structures have been utilized in characterizing the efficacy of small molecule anti-cancer drugs. In this study, we performed single molecule measurements to study how stability of GQ against Bloom helicase (BLM) is impacted by its interactions with several prominent small molecules (L1H1-7OTD, PDS and PhenDC3). We considered two different approaches to quantify this effect: dynamic approach and steady state approach. DNA constructs were accordingly designed depending on the question we addressed in each approach. The dynamic assay with time resolution of 50 ms reveals ∼50%-60% reduction in the GQ unfolding rate when a small molecule is present. On the other hand, the fraction of GQ molecules that remain folded against BLM-mediated unfolding rose by a factor of two in the presence of these small molecules. The similarity of the results attained from the two approaches illustrates a consistent picture in terms of the impact of these molecules on BLM-GQ interactions. Moreover, we investigated the impact of these molecules on folding dynamics of GQ. Even though GQ folding was too fast to characterize the impact of small molecules, we observed a significant variation in the folding conformation in the presence of small molecules.

Single-Molecule Force Spectroscopy of M-Value Mutants of Staphylococcal Nuclease Indicates Complex Protein Folding Landscape

Point mutations have commonly been utilized to investigate the underlying thermodynamic characteristics of proteins. While denaturant-induced unfolding can provide a quantitative measure for the stability of a protein, interpretation of the results depends on the model with which the equations are built. Derived from the analysis, a m-value relates the susceptibility of a protein to a denaturant and can be used as a proxy to the change in solvent-accessible surface area between the native and denatured states. In the case of the model protein Staphylococcal Nuclease (SNase), many mutants that alter the m-value have been identified. There is a long-standing debate as to whether the mutations favor an intermediate state in the unfolding pathway or perturb the residual structure of the denatured state of the protein. Bulk experiments have historically been used to try to discriminate between the various models, but these methods are insensitive to finer details of the unfolding/folding pathway(s) due to signal averaging. Single-molecule force spectroscopy via optical tweezers in the presence of chemical denaturant offers an alternative approach by following individual trajectories of the unfolding/folding pathway(s). Our force ramp experiments have revealed the presence of an intermediate in both the wild-type and mutants of SNase, which was confirmed in subsequent force clamp experiments. Further experimentation using optical tweezers allowed the determination of unfolding rates and contour length changes associated with the folding/unfolding pathway of SNase.

Unfolding Dynamics of Ubiquitin from Constant Force MD Simulation: Entropy-Enthalpy Interplay Shapes the Free-Energy Landscape.

Force probe methods are routinely used to study conformational transitions of biomolecules at single-molecule level. In contrast to simple kinetics, some proteins show complex response to mechanical perturbations that is manifested in terms of unusual force-dependent kinetics. Here, we study, via fully atomistic molecular dynamics simulations, constant force-induced unfolding of ubiquitin protein. Our simulations reveal a crossover at an intermediate force (about 400 pN) in the unfolding rate versus force curve. We find by calculation of multidimensional free-energy landscape (FEL) of the protein that the complex unfolding kinetics is intimately related to the force-dependent modifications in the FEL. Pearson correlation coefficient analysis allowed us to identify two appropriate order parameters describing the unfolding transition. The crossover in the rate can be explained in terms of an interplay between entropy and enthalpy with relative importance changing from low force to high force. We rationalize the results by using multidimensional transition-state theory.

DNA i-Motifs With Guanidino-i-Clamp Residues: The Counterplay Between Kinetics and Thermodynamics and Implications for the Design of pH Sensors

I-motif structures, adopted by cytosine-rich DNA strands, have attracted considerable interest as possible regulatory elements in genomes. Applied science exploits the advantages of i-motif stabilization under acidic conditions: i-motif-based pH sensors and other biocompatible nanodevices are being developed. Two key characteristics of i-motifs as core elements of nanodevices, i.e., their stability under physiological conditions and folding/unfolding rates, still need to be improved. We have previously reported a phenoxazine derivative (i-clamp) that enhances the thermal stability of the i-motif and shifts the pH transition point closer to physiological values. Here, we performed i-clamp guanidinylation to further explore the prospects of clamp-like modifications in i-motif fine-tuning. Based on molecular modeling data, we concluded that clamp guanidinylation facilitated interstrand interactions in an i-motif core and ultimately stabilized the i-motif structure. We tested the effects of guanidino-i-clamp insertions on the thermal stabilities of genomic and model i-motifs. We also investigated the folding/unfolding kinetics of native and modified i-motifs under moderate, physiologically relevant pH alterations. We demonstrated fast folding/unfolding of native genomic and model i-motifs in response to pH stimuli. This finding supports the concept of i-motifs as possible genomic regulatory elements and encourages the future design of rapid-response pH probes based on such structures. Incorporation of guanidino-i-clamp residues at/near the 5′-terminus of i-motifs dramatically decreased the apparent unfolding rates and increased the thermal stabilities of the structures. This counterplay between the effects of modifications on i-motif stability and their effects on kinetics should be taken into account in the design of pH sensors.

Mispacking of the Phe87 Side Chain Reduces the Kinetic Stability of Human Transthyretin.

Aggregation of transthyretin (TTR) causes TTR amyloidoses. The native TTR tetramer (a dimer of dimers) is stabilized by packing of phenylalanine 87 (F87) into a hydrophobic cavity of a neighboring protomer across the strong dimer interface. X-ray structures at acidic pH show that the side chain of F87 can be displaced from its binding pocket, but the resultant solution conformations remain unknown. Here we used 19F nuclear magnetic resonance (NMR) and 19F-labeled C10S-S85C TTR to characterize two local conformations of the loop containing F87. At neutral pH, F87 packs correctly into the interprotomer cavity in the dominant conformational state (93% population, T) whereas the remaining minor population is a mispacked tetramer (T*). The population of T* can be enhanced in heterotetramers by mixing C10S-S85C TTR with increasing molar ratios of A120L TTR, where a bulky leucine residue is introduced to disfavor the T state by steric hindrance. Exchange between the T and T* states in the presence of A120L is mediated by subunit exchange from the C10S-S85C tetramer. Compared to the TTR tetramer in which the dimers are correctly packed, mispacking of one or both dimer pairs leads to an increase in the urea unfolding rate of 4-fold or at least 15-fold, respectively. Consistent acid-mediated tetramer dissociation was observed by 19F NMR aggregation assays. Our results highlight the important role of the interprotomer F87 side chain packing in determining the kinetic stability of the TTR tetramer; mispacking of F87 in the T* state predisposes it for rapid dissociation and entry into the aggregation pathway.

Exploring the effects of cosolutes and crowding on the volumetric and kinetic profile of the conformational dynamics of a poly dA loop DNA hairpin: a single-molecule FRET study.

We investigated the volumetric and kinetic profile of the conformational landscape of a poly dA loop DNA hairpin (Hp) in the presence of salts, osmolytes and crowding media, mimicking the intracellular milieu, using single-molecule FRET methodology. Pressure modulation was applied to explore the volumetric and hydrational characteristics of the free-energy landscape of the DNA Hp, but also because pressure is a stress factor many organisms have to cope with, e.g. in the deep sea where pressures even up to the kbar level are encountered. Urea and pressure synergistically destabilize the closed conformation of the DNA Hp due to a lower molar partial volume in the unfolded state. Conversely, multivalent salts, trimethylamine-N-oxide and Ficoll strongly populate the closed state and counteract deteriorating effects of pressure. Complementary smFRET measurements under immobilized conditions at ambient pressure allowed us to dissect the equilibrium data in terms of folding and unfolding rate constants of the conformational transitions, leading to a deeper understanding of the stabilization mechanisms of the cosolutes. Our results show that the free-energy landscape of the DNA Hp is a rugged one, which is markedly affected by the ionic strength of the solution, by preferential interaction and exclusion of cosolvents as well as by pressure.

Amino Acid Stabilization of Nucleic Acid Secondary Structure: Kinetic Insights from Single-Molecule Studies.

Amino acid and nucleic acid interactions are central in biology and may have played a role in the evolutionary development of protein-based life from an early "RNA Universe." To explore the possible role of single amino acids in promoting nucleic acid folding, single-molecule Förster resonance energy transfer experiments have been implemented with a DNA hairpin construct (7 nucleotide double strand with a 40A loop) as a simple model for secondary structure formation. Exposure to positively charged amino acids (arginine and lysine) is found to clearly stabilize the secondary structure. Kinetically, each amino acid promotes folding by generating a large increase in the folding rate with little change in the unfolding rate. From analysis as a function of temperature, arginine and lysine are found to significantly increase the overall exothermicity of folding while imposing only a small entropic penalty on the folding process. Detailed investigations into the kinetics and thermodynamics of this amino acid-induced folding stability reveal arginine and lysine to interact with nucleic acids in a manner reminiscent of monovalent cations. Specifically, these observations are interpreted in the context of an ion atmosphere surrounding the nucleic acid, in which amino acid salts stabilize folding qualitatively like small monovalent cations but also exhibit differences because of the composition of their side chains.

Membrane curvature affects the stability and folding kinetics of bacteriorhodopsin

Biological membrane is crucial for the function, stability and folding of membrane proteins. By studying the stability and folding kinetics of bacteriorhodopsin (bR) in lipid vesicles with different sizes, here we report the influence of membrane curvature (vesicle size) on the stability and folding kinetics of bR. The results show that both the stability and folding kinetics of bR can be significantly changed when reconstituted into mimic membranes with different curvatures. The stability of bR decreases and unfolding rate of bR increases with the growth of vesicle size, i.e. decrease of membrane curvature. Our results suggest that it is possible to regulate the properties of membrane proteins by changing the curvature of membranes.

The Effect of Electrostatic Interactions on the Folding Kinetics of a 3-α-Helical Bundle Protein Family.

The trio of protein segment repeats called spectrins diverges by more than 2 orders of magnitude in their folding and unfolding rates, despite having very similar stabilities and almost coincidental topologies. Experimental studies revealed that the mutation of five particular residues dramatically alters the kinetic rates in the slow folders, making them similar to the rates of the fast folder. This is considered to be an exceptional behavior which seems in principle to challenge the current understanding of the protein folding process. In this work, we analyze this scenario, using a simplified computational model, combined with state-of-the-art kinetic analysis techniques. Our model faithfully separates the kinetics of the fast and slow folders and captures the effect of the five mutations. We show that the inclusion of electrostatics in the model is necessary to explain the experimental findings.