Phenotypic screening of the ReFRAME drug repurposing library to discover new drugs for treating sickle cell disease.

Stem cell transplantation and genetic therapies offer potential cures for patients with sickle cell disease (SCD), but these options require advanced medical facilities and are expensive. Consequently, these treatments will not be available for many years to the majority of patients suffering from this disease. What is urgently needed now is an inexpensive oral drug in addition to hydroxyurea, the only drug approved by the FDA that inhibits sickle-hemoglobin polymerization. Here, we report the results of the first phase of our phenotypic screen of the 12,657 compounds of the Scripps ReFRAME drug repurposing library using a recently developed high-throughput assay to measure sickling times following deoxygenation to 0% oxygen of red cells from sickle trait individuals. The ReFRAME library is a very important collection because the compounds are either FDA-approved drugs or have been tested in clinical trials. From dose-response measurements, 106 of the 12,657 compounds exhibit statistically significant antisickling at concentrations ranging from 31 nM to 10 µM. Compounds that inhibit sickling of trait cells are also effective with SCD cells. As many as 21 of the 106 antisickling compounds emerge as potential drugs. This estimate is based on a comparison of inhibitory concentrations with free concentrations of oral drugs in human serum. Moreover, the expected therapeutic potential for each level of inhibition can be predicted from measurements of sickling times for cells from individuals with sickle syndromes of varying severity. Our results should motivate others to develop one or more of these 106 compounds into drugs for treating SCD.

Allosteric control of hemoglobin S fiber formation by oxygen and its relation to the pathophysiology of sickle cell disease.

The pathology of sickle cell disease is caused by polymerization of the abnormal hemoglobin S upon deoxygenation in the tissues to form fibers in red cells, causing them to deform and occlude the circulation. Drugs that allosterically shift the quaternary equilibrium from the polymerizing T quaternary structure to the nonpolymerizing R quaternary structure are now being developed. Here we update our understanding on the allosteric control of fiber formation at equilibrium by showing how the simplest extension of the classic quaternary two-state allosteric model of Monod, Wyman, and Changeux to include tertiary conformational changes provides a better quantitative description. We also show that if fiber formation is at equilibrium in vivo, the vast majority of cells in most tissues would contain fibers, indicating that it is unlikely that the disease would be survivable once the nonpolymerizing fetal hemoglobin has been replaced by adult hemoglobin S at about 1 y after birth. Calculations of sickling times, based on a recently discovered universal relation between the delay time prior to fiber formation and supersaturation, show that in vivo fiber formation is very far from equilibrium. Our analysis indicates that patients survive because the delay period allows the majority of cells to escape the small vessels of the tissues before fibers form. The enormous sensitivity of the duration of the delay period to intracellular hemoglobin composition also explains why sickle trait, the heterozygous condition, and the compound heterozygous condition of hemoglobin S with pancellular hereditary persistence of fetal hemoglobin are both relatively benign conditions.

Kinetic assay shows that increasing red cell volume could be a treatment for sickle cell disease.

Although it has been known for more than 60 years that the cause of sickle cell disease is polymerization of a hemoglobin mutant, hydroxyurea is the only drug approved for treatment by the US Food and Drug Administration. This drug, however, is only partially successful, and the discovery of additional drugs that inhibit fiber formation has been hampered by the lack of a sensitive and quantitative cellular assay. Here, we describe such a method in a 96-well plate format that is based on laser-induced polymerization in sickle trait cells and robust, automated image analysis to detect the precise time at which fibers distort ("sickle") the cells. With this kinetic method, we show that small increases in cell volume to reduce the hemoglobin concentration can result in therapeutic increases in the delay time prior to fiber formation. We also show that, of the two drugs (AES103 and GBT440) in clinical trials that inhibit polymerization by increasing oxygen affinity, one of them (GBT440) also inhibits sickling in the absence of oxygen by two additional mechanisms.

Making connections between ultrafast protein folding kinetics and molecular dynamics simulations.

Determining the rate of forming the truly folded conformation of ultrafast folding proteins is an important issue for both experiments and simulations. The double-norleucine mutant of the 35-residue villin subdomain is the focus of recent computer simulations with atomistic molecular dynamics because it is currently the fastest folding protein. The folding kinetics of this protein have been measured in laser temperature-jump experiments using tryptophan fluorescence as a probe of overall folding. The conclusion from the simulations, however, is that the rate determined by fluorescence is significantly larger than the rate of overall folding. We have therefore employed an independent experimental method to determine the folding rate. The decay of the tryptophan triplet-state in photoselection experiments was used to monitor the change in the unfolded population for a sequence of the villin subdomain with one amino acid difference from that of the laser temperature-jump experiments, but with almost identical equilibrium properties. Folding times obtained in a two-state analysis of the results from the two methods at denaturant concentrations varying from 1.5-6.0 M guanidinium chloride are in excellent agreement, with an average difference of only 20%. Polynomial extrapolation of all the data to zero denaturant yields a folding time of 220 (+100,-70) ns at 283 K, suggesting that under these conditions the barrier between folded and unfolded states has effectively disappeared--the so-called "downhill scenario."

Measuring internal friction of an ultrafast-folding protein.

Nanosecond laser T-jump was used to measure the viscosity dependence of the folding kinetics of the villin subdomain under conditions where the viscogen has no effect on its equilibrium properties. The dependence of the unfolding/refolding relaxation time on solvent viscosity indicates a major contribution to the dynamics from internal friction. The internal friction increases with increasing temperature, suggesting a shift in the transition state along the reaction coordinate toward the native state with more compact structures, and therefore, a smaller diffusion coefficient due to increased landscape roughness. Fitting the data with an Ising-like model yields a relatively small position dependence for the diffusion coefficient. This finding is consistent with the excellent correlation found between experimental and calculated folding rates based on free energy barrier heights using the same diffusion coefficient for every protein.

Chemical, physical, and theoretical kinetics of an ultrafast folding protein.

An extensive set of equilibrium and kinetic data is presented and analyzed for an ultrafast folding protein--the villin subdomain. The equilibrium data consist of the excess heat capacity, tryptophan fluorescence quantum yield, and natural circular-dichroism spectrum as a function of temperature, and the kinetic data consist of time courses of the quantum yield from nanosecond-laser temperature-jump experiments. The data are well fit with three kinds of models--a three-state chemical-kinetics model, a physical-kinetics model, and an Ising-like theoretical model that considers 10(5) possible conformations (microstates). In both the physical-kinetics and theoretical models, folding is described as diffusion on a one-dimensional free-energy surface. In the physical-kinetics model the reaction coordinate is unspecified, whereas in the theoretical model, order parameters, either the fraction of native contacts or the number of native residues, are used as reaction coordinates. The validity of these two reaction coordinates is demonstrated from calculation of the splitting probability from the rate matrix of the master equation for all 10(5) microstates. The analysis of the data on site-directed mutants using the chemical-kinetics model provides information on the structure of the transition-state ensemble; the physical-kinetics model allows an estimate of the height of the free-energy barrier separating the folded and unfolded states; and the theoretical model provides a detailed picture of the free-energy surface and a residue-by-residue description of the evolution of the folded structure, yet contains many fewer adjustable parameters than either the chemical- or physical-kinetics models.

Evidence for a partially structured state of the amylin monomer.

Islet amyloid polypeptide (amylin) is the main component in amyloid deposits formed in type II diabetes. We used triplet quenching to probe the dynamics of contact formation between the N-terminal disulfide loop and a C-terminal tryptophan in monomeric amylins from human and rat. Quenching rates measured in the absence of denaturant are four times larger than those in 6 M guanidinium chloride, indicating a decrease in the average end-to-end distance (collapse) at low denaturant concentrations. We were surprised to find an even greater (sevenfold) increase in quenching rates on removal of denaturant for a hydrophilic control peptide containing the disulfide loop compared to the same peptide without the loop (twofold change). These results suggest that collapse is driven by backbone-backbone and backbone-side chain interactions involving the disulfide loop portion of the chain rather than by the formation of side-chain hydrophobic contacts. Molecular dynamics simulations of the control peptide show that the collapse results from hydrogen-bonding interactions between the central residues of the chain and the disulfide loop. The quenching experiments also indicate that the monomer of the human, amyloidogenic form of amylin is more compact than the rat form, which does not form amyloid. We discuss these newly observed differences between human and rat amylin in solution and their possible relation to aggregation and to the physiological function of amylin binding to the calcitonin receptor.

Sedimentation studies on human amylin fail to detect low-molecular-weight oligomers.

Sedimentation velocity experiments show that only monomers coexist with amyloid fibrils of human islet amyloid-polypeptide. No oligomers containing <100 monomers could be detected, suggesting that the putative toxic oligomers are much larger than those found for the Alzheimer's peptide, Abeta(1-42).

Estimating free-energy barrier heights for an ultrafast folding protein from calorimetric and kinetic data.

Differential scanning calorimetry was used to measure the temperature dependence of the absolute heat capacity of the 35-residue subdomain of the villin headpiece, a protein that folds in 5 mus and is therefore assumed to have a small free-energy barrier separating folded and unfolded states. To obtain an estimate of the barrier height from the calorimetric data, two models, a variable-barrier model and an Ising-like model, were used to fit the heat capacity in excess of the folded state over the temperature range 15-125 degrees C. The variable-barrier model is based on an empirical mathematical form for the density of states, with four adjustable parameters and the enthalpy (H) as a reaction coordinate. The Ising-like model is based on the inter-residue contact map of the X-ray structure with exact enumeration of approximately 10(5) possible conformations, with two adjustable parameters in the partition function, and either the fraction of native contacts (Q) or the number of ordered residues (P) as reaction coordinates. The variable-barrier model provides an excellent fit to the data and yields a barrier height at the folding temperature ranging from 0.4 to 1.1 kcal mol(-1), while the Ising-like model provides a less good fit and yields barrier heights of 2.3 +/- 0.1 kcal mol(-1) and 2.1 +/- 0.1 kcal mol(-1) for the Q and P reaction coordinates, respectively. In both models, the barrier to folding increases with increasing temperature. Assuming a sufficiently large activation energy for diffusion on the free-energy surfaces, both models are consistent with the observation of a temperature-independent folding rate in previously published laser temperature-jump experiments. Analysis of this kinetic data, using an approximate form for the pre-exponential factor of Kramers theory and the 70 ns relaxation time for the fast phase that precedes the unfolding/refolding relaxation to determine the diffusion coefficient, results in a barrier height of 1.6 +/- 0.3 kcal mol-1 for an unspecified reaction coordinate. Although no independent test of the validity of the H, Q, or P reaction coordinates is given, the barrier-height estimates obtained with the three reaction coordinates are in quite good agreement with the value derived from a Kramers analysis of the kinetics that makes no assumptions about the reaction coordinate. However, the higher estimates obtained using Q or P appear more consistent with the finding of barrier-crossing kinetics of a villin mutant that folds in 700 ns, corresponding to a 1.3 kcal mol-1 reduction in the folding barrier relative to wild-type. All of the results suggest that the free-energy barrier to folding is sufficiently low that it should be possible to engineer this protein or find solution conditions that would eliminate the barrier to create the "downhill" folding scenario of Wolynes and Onuchic.

Relaxation rate for an ultrafast folding protein is independent of chemical denaturant concentration.

The connection between free-energy surfaces and chevron plots has been investigated in a laser temperature jump kinetic study of a small ultrafast folding protein, the 35-residue subdomain from the villin headpiece. Unlike all other proteins that have been studied so far, no measurable dependence of the unfolding/refolding relaxation rate on denaturant concentration was observed over a wide range of guanidinium chloride concentration. Analysis with a simple Ising-like theoretical model shows that this denaturant-invariant relaxation rate can be explained by a large movement of the major free energy barrier, together with a denaturant- and reaction coordinate-dependent diffusion coefficient.

Sub-microsecond protein folding.

We have investigated the structure, equilibria, and folding kinetics of an engineered 35-residue subdomain of the chicken villin headpiece, an ultrafast-folding protein. Substitution of two buried lysine residues by norleucine residues stabilizes the protein by 1 kcal/mol and increases the folding rate sixfold, as measured by nanosecond laser T-jump. The folding rate at 300 K is (0.7 micros)(-1) with little or no temperature dependence, making this protein the first sub-microsecond folder, with a rate only twofold slower than the theoretically predicted speed limit. Using the 70 ns process to obtain the effective diffusion coefficient, the free energy barrier height is estimated from Kramers theory to be less than approximately 1 kcal/mol. X-ray crystallographic determination at 1A resolution shows no significant change in structure compared to the single-norleucine-substituted molecule and suggests that the increased stability is electrostatic in origin. The ultrafast folding rate, very accurate X-ray structure, and small size make this engineered villin subdomain an ideal system for simulation by atomistic molecular dynamics with explicit solvent.

The protein folding 'speed limit'.

How fast can a protein possibly fold? This question has stimulated experimentalists to seek fast folding proteins and to engineer them to fold even faster. Proteins folding at or near the speed limit are prime candidates for all-atom molecular dynamics simulations. They may also have no free energy barrier, allowing the direct observation of intermediate structures on the pathways from the unfolded to the folded state. Both experimental and theoretical approaches predict a speed limit of approximately N/100micros for a generic N-residue single-domain protein, with alpha proteins folding faster than beta or alphabeta. The predicted limits suggest that most known ultrafast folding proteins can be engineered to fold more than ten times faster.

Determination of ultrafast protein folding rates from loop formation dynamics.

Quenching of the triplet state of tryptophan by contact with cysteine can be used to measure the kinetics of loop formation in unfolded proteins. Here we show that cysteine quenching dynamics also provide a novel method for measuring folding rates when the exchange between folded and unfolded states is faster than the unquenched triplet lifetime (approximately 100 micros). We use this technique to investigate folding/unfolding kinetics of the 35 residue headpiece subdomain of the protein villin, which contains a single tryptophan residue and was engineered to contain a cysteine residue at the N terminus. At intermediate concentrations of denaturant the time-course of the triplet decay consists of two relaxations, the rates and amplitudes of which reveal the fast kinetics for folding and unfolding of this protein. The folding rates extracted using a simple kinetic model are close to those reported previously from laser-induced temperature-jump experiments that employ the change in tryptophan fluorescence as a probe. However, the results differ significantly from those reported from dynamic NMR line shape analysis on a variant with methionine at the N terminus, an issue that remains to be resolved. The analysis of the triplet quenching kinetics also shows that the quenching rates in the unfolded state increase with decreasing denaturant concentration, indicating a compaction of the unfolded protein.

Measuring dynamic flexibility of the coil state of a helix-forming peptide.

To investigate the dynamic flexibility of the coil state of a helix-forming peptide the end-to-end contact rate was determined. Nanosecond optical excitation of tryptophan at one end of a 22 residue, alanine peptide populates a long-lived triplet state which is quenched upon close contact with a cyclic disulfide attached to the opposite end. Analysis of the decay of the triplet population using a two-state model for helix formation yields the diffusion-limited end-to-end contact rate of the coil state of the peptide as well as the helix-->coil and coil-->helix rates. The helix-coil rates are very similar to those previously measured in laser temperature-jump experiments. The end-to-end contact rate of 1.1 x 10(7) s(-1) in the coil state is tenfold faster than the rate for a disordered peptide with threonine substituted for alanine and, somewhat surprisingly, is about twice the rate for a disordered glycine-containing peptide. These differences are discussed in terms of the theory of Szabo, Schulten and Schulten. The rates should provide important new benchmarks for testing the accuracy of atomistic molecular dynamics simulations.

Experimental tests of villin subdomain folding simulations.

We have used laser temperature-jump to investigate the kinetics and mechanism of folding the 35 residue subdomain of the villin headpiece. The relaxation kinetics are biphasic with a sub-microsecond phase corresponding to a helix-coil transition and a slower microsecond phase corresponding to overall unfolding/refolding. At 300 K, the folding time is 4.3(+/-0.6) micros, making it the fastest folding, naturally occurring protein, with a rate close to the theoretical speed limit. This time is in remarkable agreement with the prediction of 5 (+11,-3) micros by Zagrovic et al. from atomistic molecular dynamics simulations using an implicit solvent model. We test their prediction that replacement of the C-terminal phenylalanine residue with alanine will increase the folding rate by removing a transient non-native interaction. We find that the alanine substitution has no effect on the folding rate or on the equilibrium constant. Implications of this result for the validity of the simulated folding mechanism are discussed.

Kinetics of intramolecular contact formation in a denatured protein.

Quenching of the triplet state of tryptophan by cysteine has provided a new tool for measuring the rate of forming a specific intramolecular contact in disordered polypeptides. Here, we use this technique to investigate contact formation in the denatured state of CspTm, a small cold-shock protein from Thermotoga maritima, engineered to contain a single tryptophan residue (W29) and a single cysteine residue at the C terminus (C67). At all concentrations of denaturant, the decay rate of the W29 triplet of the unfolded protein is more than tenfold faster than the rate observed for the native protein ( approximately 10(4)s(-1)). Experiments on the unfolded protein without the added C-terminal cysteine residue show that this faster rate results entirely from contact quenching by C67. The quenching rate in the unfolded state by C67 increases at concentrations of denaturant that favor folding, indicating a compaction of the unfolded protein as observed previously in single-molecule Förster resonance energy transfer (FRET) experiments.

A tertiary two-state allosteric model for hemoglobin.

The two-state allosteric model of Monod, Wyman, and Changeux (MWC) provides an excellent description of homotropic effects in a vast array of equilibrium and kinetic measurements on cooperative ligand binding by hemoglobin. However, in contrast to experimental observations, the model does not allow for alteration of the ligand affinity of the T quaternary structure by allosteric effectors. This failure to explain heterotropic effects has been appreciated for over 30 years, and it has been generally assumed to result from tertiary conformational changes in the absence of a quaternary change. Here we explore a model that preserves the essential MWC idea that binding without a quaternary conformational change is non-cooperative, but where tertiary conformations of individual subunits play the primary role instead of the quaternary conformations. In this model, which we call the 'tertiary two-state (TTS) model', the two affinity states correspond to two tertiary conformations of individual subunits rather than the two quaternary conformations of the MWC two-state allosteric model. Ligation and the R quaternary structure bias the subunit population toward the high affinity tertiary conformation, while deligation and the T quaternary structure favor the low affinity tertiary conformation. We show that the model is successful in quantitatively explaining a demanding set of kinetic data from nanosecond carbon monoxide photodissociation experiments at times longer than approximately 1 micros. Better agreement between the model and the submicrosecond kinetic data may result from detailed considerations of the distribution and dynamics of conformational substates of the two tertiary conformations. The model is consistent with the results of solution, gel, and single crystal oxygen binding studies, but underestimates the population of doubly-liganded molecules determined in low-temperature electrophoresis experiments.

Evolution of allosteric models for hemoglobin.

We compare various allosteric models that have been proposed to explain cooperative oxygen binding to hemoglobin, including the two-state allosteric model of Monod, Wyman, and Changeux (MWC), the Cooperon model of Brunori, the model of Szabo and Karplus (SK) based on the stereochemical mechanism of Perutz, the generalization of the SK model by Lee and Karplus (SKL), and the Tertiary Two-State (TTS) model of Henry, Bettati, Hofrichter and Eaton. The preponderance of experimental evidence favors the TTS model which postulates an equilibrium between high (r)- and low (t)-affinity tertiary conformations that are present in both the T and R quaternary structures. Cooperative oxygenation in this model arises from the shift of T to R, as in MWC, but with a significant population of both r and t conformations in the liganded T and in the unliganded R quaternary structures. The TTS model may be considered a combination of the SK and SKL models, and these models provide a framework for a structural interpretation of the TTS parameters. The most compelling evidence in favor of the TTS model is the nanosecond - millisecond carbon monoxide (CO) rebinding kinetics in photodissociation experiments on hemoglobin encapsulated in silica gels. The polymeric network of the gel prevents any tertiary or quaternary conformational changes on the sub-second time scale, thereby permitting the subunit conformations prior to CO photodissociation to be determined from their ligand rebinding kinetics. These experiments show that a large fraction of liganded subunits in the T quaternary structure have the same functional conformation as liganded subunits in the R quaternary structure, an experimental finding inconsistent with the MWC, Cooperon, SK, and SKL models, but readily explained by the TTS model as rebinding to r subunits in T. We propose an additional experiment to test another key prediction of the TTS model, namely that a fraction of subunits in the unliganded R quaternary structure has the same functional conformation (t) as unliganded subunits in the T quaternary structure.

Effects of denaturants on the dynamics of loop formation in polypeptides.

Quenching of the triplet state of tryptophan by close contact with cysteine has been used to measure the reaction-limited and diffusion-limited rates of loop formation in disordered polypeptides having the sequence cys-(ala-gly-gln)j-trp (j=1-9). The decrease in the length-dependence of the reaction-limited rate for short chains in aqueous buffer, previously attributed to chain stiffness, is not observed at high concentrations of chemical denaturant (6 M GdmCl and 8 M urea), showing that denaturants increase chain flexibility. For long chains, both reaction-limited and diffusion-limited rates are significantly smaller in denaturant and exhibit a steeper length dependence. The results can be explained using end-to-end distributions from a wormlike chain model in which excluded volume interactions are incorporated by associating a 0.4-0.5 nm diameter hard sphere with the end of each virtual peptide bond. Fitting the data with this model shows that the denaturants reduce the persistence length from approximately 0.6 nm to approximately 0.4 nm, only slightly greater than the length of a peptide bond. The same model also describes the reported length dependence for the radii of gyration of chemically denatured proteins containing 50-400 residues. The end-to-end diffusion coefficients obtained from the diffusion-limited rates are smaller than the sum of the monomer diffusion coefficients and exhibit significant temperature dependence, suggesting that diffusion is slowed by internal friction arising from barriers to backbone conformational changes.

Folding and aggregation kinetics of a beta-hairpin.

We have investigated the solution structure, equilibrium properties, and folding kinetics of a 17-residue beta-hairpin-forming peptide derived from the protein ubiquitin. NMR experiments show that at 4 degrees C the peptide has a highly populated beta-hairpin conformation. At protein concentrations higher than 0.35 mM, the peptide aggregates. Sedimentation equilibrium measurements show that the aggregate is a trimer, while NMR indicates that the beta-hairpin conformation is maintained in the trimer. The relaxation kinetics in nanosecond laser temperature-jump experiments reveal a concentration-independent microsecond phase, corresponding to beta-hairpin unfolding-refolding, and a concentration-dependent millisecond phase due to oligomerization. Kinetic modeling of the relaxation rates and amplitudes yields the folding and unfolding rates for the monomeric beta-hairpin, as well as assembly and disassembly rates for trimer formation consistent with the equilibrium constant determined by sedimentation equilibrium. When the net charge on the peptides and ionic strength were taken into account, the rate of trimer assembly approaches the Debye-Smoluchowski diffusion limit. At 300 K, the rate of formation of the monomeric hairpin is (17 micros)(-1), compared to rates of (0.8 micros)(-1) to (52 micros)(-1) found for other peptides. After using Kramers theory to correct for the temperature dependence of the pre-exponential factor, the activation energy for hairpin formation is near zero, indicating that the barrier to folding is purely entropic. Comparisons with previously measured rates for a series of hairpins are made to distinguish between zipper and hydrophobic collapse mechanisms. Overall, the experimental data are most consistent with the zipper mechanism in which structure formation is initiated at the turn, the mechanism predicted by the Ising-like statistical mechanical model that was developed to explain the equilibrium and kinetic data for the beta-hairpin from protein GB1. In contrast, the majority of simulation studies favor a hydrophobic collapse mechanism. However, with few exceptions, there is little or no quantitative comparison of the simulation results with experimental data.