Publications

Quantifying biomolecular dynamics has become a major task of single-molecule fluorescence spectroscopy methods. In single-molecule Förster resonance energy transfer (smFRET), kinetic information is extracted from the stream of photons emitted by attached donor and acceptor fluorophores. Here, we describe a time-resolved version of burst variance analysis that can quantify kinetic rates at microsecond to millisecond timescales in smFRET experiments of diffusing molecules. Bursts are partitioned into segments with a fixed number of photons. The FRET variance is computed from these segments and compared with the variance expected from shot noise. By systematically varying the segment size, dynamics at different timescales can be captured. We provide a theoretical framework to extract kinetic rates from the decay of the FRET variance with increasing segment size. Compared to other methods such as filtered fluorescence correlation spectroscopy, recurrence analysis of single particles, and two-dimensional lifetime correlation spectroscopy, fewer photons are needed to obtain reliable timescale estimates, which reduces the required measurement time.

Intrinsically disordered proteins often form dynamic complexes with their ligands. Yet, the speed and amplitude of these motions are hidden in classical binding kinetics. Here, we directly measure the dynamics in an exceptionally mobile, high-affinity complex. We show that the disordered tail of the cell adhesion protein E-cadherin dynamically samples a large surface area of the protooncogene β-catenin. Single-molecule experiments and molecular simulations resolve these motions with high resolution in space and time. Contacts break and form within hundreds of microseconds without a dissociation of the complex. The energy landscape of this complex is rugged with many small barriers (3 to 4 kBT) and reconciles specificity, high affinity, and extreme disorder. A few persistent contacts provide specificity, whereas unspecific interactions boost affinity.

The thermal motion of charged proteins causes randomly fluctuating electric fields inside cells. According to the fluctuation-dissipation theorem, there is an additional friction force associated with such fluctuations. However, the impact of these fluctuations on the diffusion and dynamics of proteins in the cytoplasm is unclear. Here, we provide an order-of-magnitude estimate of this effect by treating electric field fluctuations within a generalized Langevin equation model with a time-dependent friction memory kernel. We find that electric friction is generally negligible compared to solvent friction. However, a significant slowdown of protein diffusion and dynamics is expected for biomolecules with high net charges such as intrinsically disordered proteins and RNA. The results show that direct contacts between biomolecules in a cell are not necessarily required to alter their dynamics.

The microtubule-associated protein, tau, is the major subunit of neurofibrillary tangles associated with neurodegenerative conditions, such as Alzheimer's disease. In the cell, however, tau aggregation can be prevented by a class of proteins known as molecular chaperones. While numerous chaperones are known to interact with tau, though, little is known regarding the mechanisms by which these prevent tau aggregation. Here, we describe the effects of ATP-independent Hsp40 chaperones, DNAJA2 and DNAJB1, on tau amyloid-fiber formation, and compare these to the small heat-shock protein HSPB1. We find that the chaperones play complementary roles, with each preventing tau aggregation differently and interacting with distinct sets of tau species. Whereas HSPB1 only binds tau monomers, DNAJB1 and DNAJA2 recognize aggregation-prone conformers and even mature fibers. In addition, we find that both Hsp40s bind tau seeds and fibers via their C-terminal domain II (CTDII), with DNAJA2 being further capable of recognizing tau monomers by a second, distinct site in CTDI. These results lay out the mechanisms by which the diverse members of the Hsp40 family counteract the formation and propagation of toxic tau aggregates, and highlight the fact that chaperones from different families/classes play distinct, yet complementary roles in preventing pathological protein aggregation.

Structural disorder is widespread in regulatory protein networks. Weak and transient interactions render disordered proteins particularly sensitive to fluctuations in solution conditions such as ion and crowder concentrations. How this sensitivity alters folding coupled binding reactions, however, has not been fully understood. Here, we demonstrate that salt jointly modulates polymer properties and binding affinities of 5 disordered proteins from a transcription factor network. A combination of single-molecule Forster resonance energy transfer experiments, polymer theory, and molecular simulations shows that all 5 proteins expand with increasing ionic strengths due to Debye-Htickel charge screening. Simultaneously, pairwise affinities between the proteins increase by an order of magnitude within physiological salt limits. A quantitative analysis shows that 50% of the affinity increase can be explained by changes in the disordered state. Disordered state properties therefore have a functional relevance even if these states are not directly involved in biological functions. Numerical solutions of coupled binding equilibria with our results show that networks of homologous disordered proteins can function surprisingly robustly in fluctuating cellular environments, despite the sensitivity of its individual proteins.

Quantifying the stability of intermediates along parallel molecular pathways is often hampered by the limited experimental resolution of ensemble methods. In biology, however, such intermediates may represent important regulatory targets, thus calling for strategies to map their abundance directly. Here, we use single-molecule Forster resonance energy transfer (FRET) to quantify the occupancies of intermediates along two parallel DNA-binding pathways of the basic helix-loop-helix leucine-zipper (bHLH-LZ) domains of the transcription factors c-Myc and Max. We find that both proteins are intrinsically disordered with sub-microsecond end-to-end distance dynamics. In mixtures of the proteins with their promoter DNA, our experiments identify the disordered conformers, the folded Myc-Max dimer, and ternary Myc-Max-DNA complexes. However, signatures of the intermediate along the alternative pathway, i.e., one domain bound to DNA, remained undetectable. This implies that disordered Max-DNA and Myc-DNA complexes are by at least 6 k(B)T higher in free energy than folded dimers of Myc and Max. The disordered monomer-DNA complex is therefore unlikely to be of importance for the regulation of transcriptional processes.

Protein dynamics are typically captured well by rate equations that predict exponential decays for two-state reactions. Here, we describe a remarkable exception. The electron-transfer enzyme quiescin sulfhydryl oxidase (QSOX), a natural fusion of two functionally distinct domains, switches between open- and closed-domain arrangements with apparent power-law kinetics. Using single-molecule FRET experiments on time scales from nanoseconds to milliseconds, we show that the unusual openclose kinetics results from slow sampling of an ensemble of disordered domain orientations. While substrate accelerates the kinetics, thus suggesting a substrate-induced switch to an alternative free energy landscape of the enzyme, the power-law behavior is also preserved upon electron load. Our results show that the slow sampling of open conformers is caused by a variety of interdomain interactions that imply a rugged free energy landscape, thus providing a generic mechanism for dynamic disorder in multidomain enzymes.

Single-molecule fluorescence spectroscopy is a powerful approach for probing biomolecular structure and dynamics, including protein folding. For the investigation of nonequilibrium kinetics, Förster resonance energy transfer combined with confocal multiparameter detection has proven particularly versatile, owing to the large number of observables and the broad range of accessible timescales, especially in combination with rapid microfluidic mixing. However, a comprehensive kinetic analysis of the resulting time series of transfer efficiency histograms and complementary observables can be challenging owing to the complexity of the data. Here we present and compare three different methods for the analysis of such kinetic data: singular value decomposition, multivariate curve resolution with alternating least square fitting, and model-based peak fitting, where an explicit model of both the transfer efficiency histogram of each species and the kinetic mechanism of the process is employed. While each of these methods has its merits for specific applications, we conclude that model-based peak fitting is most suitable for a quantitative analysis and comparison of kinetic mechanisms.

There has been a long-standing controversy regarding the effect of chemical denaturants on the dimensions of unfolded and intrinsically disordered proteins: A wide range of experimental techniques suggest that polypeptide chains expand with increasing denaturant concentration, but several studies using small-angle X-ray scattering (SAXS) have reported no such increase of the radius of gyration (R_g). This inconsistency challenges our current understanding of the mechanism of chemical denaturants, which are widely employed to investigate protein folding and stability. Here, we use a combination of single-molecule Förster resonance energy transfer (FRET), SAXS, dynamic light scattering (DLS), and two-focus fluorescence correlation spectroscopy (2f-FCS) to characterize the denaturant dependence of the unfolded state of the spectrin domain R17 and the intrinsically disordered protein ACTR in two different denaturants. Standard analysis of the primary data clearly indicates an expansion of the unfolded state with increasing denaturant concentration irrespective of the protein, denaturant, or experimental method used. This is the first case in which SAXS and FRET have yielded even qualitatively consistent results regarding expansion in denaturant when applied to the same proteins. To more directly illustrate this self-consistency, we used both SAXS and FRET data in a Bayesian procedure to refine structural ensembles representative of the observed unfolded state. This analysis demonstrates that both of these experimental probes are compatible with a common ensemble of protein configurations for each denaturant concentration. Furthermore, the resulting ensembles reproduce the trend of increasing hydrodynamic radius with denaturant concentration obtained by 2f-FCS and DLS. We were thus able to reconcile the results from all four experimental techniques quantitatively, to obtain a comprehensive structural picture of denaturant-induced unfolded state expansion, and to identify the most likely sources of earlier discrepancies.

Hsp90 is an energy-consuming molecular chaperone that activates oncogenic proteins in a complicated multi-step reaction. Photoinduced electron transfer (PET) quenching experiments with a fluorescent reporter have now identified molecular transitions at multiple timescales in the chaperone cycle of Hsp90.

Abstract Molecular simulation is a valuable and complementary tool that may assist with the interpretation of single-molecule Förster resonance energy transfer (FRET) experiments, if the energy function is of sufficiently high quality. Here we present force-field parameters for one of the most common pairs of chromophores used in experiments, AlexaFluor 488 and 594. From microsecond molecular-dynamics simulations, we are able to recover both experimentally determined equilibrium constants and association/dissociation rates of the chromophores with free tryptophan, as well as the decay of fluorescence anisotropy of a labeled protein. We find that it is particularly important to obtain a correct balance of solute-water interactions in the simulations in order to faithfully capture the experimental anisotropy decays, which provide a sensitive benchmark for fluorophore mobility. Lastly, by a combination of experiment and simulation, we address a potential complication in the interpretation of experiments on polyproline, used as a molecular ruler for FRET experiments, namely the potential association of one of the chromophores with the polyproline helix. Under conditions where simulations accurately capture the fluorescence anisotropy decay, we find at most a modest, transient population of conformations in which the chromophores associate with the polyproline. Explicit calculation of FRET transfer efficiencies for short polyprolines yields results in good agreement with experiment. These results illustrate the potential power of a combination of molecular simulation and experiment in quantifying biomolecular dynamics.

In the past decade, single-molecule fluorescence techniques provided important insights into the structure and dynamics of proteins. In particular, our understanding of the heterogeneous conformational ensembles of unfolded and intrinsically disordered proteins (IDPs) improved substantially by a combination of FRET-based single-molecule techniques with concepts from polymer physics. A complete knowledge of the forces that act in unfolded polypeptide chains will not only be important to understand the initial steps of protein folding reactions, but it will also be crucial to rationalize the coupling between ligand-binding and folding of IDPs, and the interaction of denatured proteins with molecular chaperones in the crowded cellular environment. Here, I give a personalized review of some of the key findings from my own research that contributed to a more quantitative understanding of unfolded proteins and their interactions with molecular chaperones.

The bacterial chaperonin GroEL/GroES assists folding of a broad spectrum of denatured and misfolded proteins. Here, we explore the limits of this remarkable promiscuity by mapping two denatured proteins with very different conformational properties, rhodanese and cyclophilin A, during binding and encapsulation by GroEL/GroES with single-molecule spectroscopy, microfluidic mixing, and ensemble kinetics. We find that both proteins bind to GroEL with high affinity in a reaction involving substantial conformational adaptation. However, whereas the compact denatured state of rhodanese is encapsulated efficiently upon addition of GroES and ATP, the more expanded and unstructured denatured cyclophilin A is not encapsulated but is expelled into solution. The origin of this surprising disparity is the weaker interactions of cyclophilin A with a transiently formed GroEL-GroES complex, which may serve as a crucial checkpoint for substrate discrimination.

The dimensions of intrinsically disordered and unfolded proteins critically depend on the solution conditions, such as temperature, pH, ionic strength, and osmolyte or denarurant concentration. However, a quantitative understanding of how the complex combination of chain-chain and chain-solvent interactions is affected by the solvent is still missing. Here, we take a step towards this goal by investigating the combined effect of pH and denaturants on the dimensions of an unfolded protein. We use single-molecule fluorescence spectroscopy to extract the dimensions of unfolded cold shock protein (CspTm) in mixtures of the denaturants urea and guanidinium chloride (GdmCl) at neutral and acidic pH. Surprisingly, even though a change in pH from 7 to 2.9 increases the net charge of CspTm from -3.8 to +10.2, the radius of gyration of the chain is very similar under both conditions, indicating that protonation of acidic side chains at low pH results in additional hydrophobic interactions. We use a simple shared binding site model that describes the joint effect of urea and GdmCl, together with polyampholyte theory and an ion cloud model that includes the chemical free energy of counterion interactions and side chain protonation, to quantify this effect.

Microfluidic mixing in combination with single-molecule spectroscopy allows the investigation of complex biomolecular processes under non-equilibrium conditions. Here we present a protocol for building, installing and operating microfluidic mixing devices optimized for this purpose. The mixer is fabricated by replica molding with polydimethylsiloxane (PDMS), which allows the production of large numbers of devices at a low cost using a single microfabricated silicon mold. The design is based on hydrodynamic focusing combined with diffusive mixing and allows single-molecule kinetics to be recorded over five orders of magnitude in time, from 1 ms to ∼100 s. Owing to microfabricated particle filters incorporated in the inlet channels, the devices provide stable flow for many hours to days without channel blockage, which allows reliable collection of high-quality data. Modular design enables rapid exchange of samples and mixing devices, which are mounted in a specifically designed holder for use with a confocal microscopy detection system. Integrated Peltier elements provide temperature control from 4 to 37C. The protocol includes the fabrication of a silicon master, production of the microfluidic devices, instrumentation setup and data acquisition. Once a silicon master is available, devices can be produced and experiments started within ∼1 d of preparation. We demonstrate the performance of the system with single-molecule Förster resonance energy transfer (FRET) measurements of kinetics of protein folding and conformational changes. The dead time of 1 ms, as predicted from finite element calculations, was confirmed by the measurements.

The dimensions of unfolded and intrinsically disordered proteins are highly dependent on their amino acid composition and solution conditions, especially salt and denaturant concentration. However, the quantitative implications of this behavior have remained unclear, largely because the effective theta-state, the central reference point for the underlying polymer collapse transition, has eluded experimental determination. Here, we used single-molecule fluorescence spectroscopy and two-focus correlation spectroscopy to determine the theta points for six different proteins. While the scaling exponents of all proteins converge to 0.62 ± 0.03 at high denaturant concentrations, as expected for a polymer in good solvent, the scaling regime in water strongly depends on sequence composition. The resulting average scaling exponent of 0.46 ± 0.05 for the four foldable protein sequences in our study suggests that the aqueous cellular milieu is close to effective theta conditions for unfolded proteins. In contrast, two intrinsically disordered proteins do not reach the Θ-point under any of our solvent conditions, which may reflect the optimization of their expanded state for the interactions with cellular partners. Sequence analyses based on our results imply that foldable sequences with more compact unfolded states are a more recent result of protein evolution.

Molecular chaperones are known to be essential for avoiding protein aggregation in vivo, but it is still unclear how they affect protein folding mechanisms. We use single-molecule Förster resonance energy transfer to follow the folding of a protein inside the GroEL/GroES chaperonin cavity over a time range from milliseconds to hours. Our results show that confinement in the chaperonin decelerates the folding of the C-terminal domain in the substrate protein rhodanese, but leaves the folding rate of the N-terminal domain unaffected. Microfluidic mixing experiments indicate that strong interactions of the substrate with the cavity walls impede the folding process, but the folding hierarchy is preserved. Our results imply that no universal chaperonin mechanism exists. Rather, a competition between intra- and intermolecular interactions determines the folding rates and mechanisms of a substrate inside the GroEL/GroES cage.

A promising approach to unravel the relationship between sequence information, tertiary structure, and folding mechanism of proteins is the analysis of the folding behavior of proteins with low sequence identity but comparable tertiary structures. Ribonuclease A (RNase A) and its homologues, forming the RNase A superfamily, provide an excellent model system for respective studies. RNase A has been used extensively as a model protein for folding studies. However, little is known about the folding of homologous RNases. Here, we analyze the folding pathway of onconase, a homologous protein from the Northern leopard frog with great potential as a tumor therapeutic, by high-resolution techniques. Although onconase and RNase A significantly differ in the primary structure (28% sequence identity) and in thermodynamic stability (ΔΔG = 20 kJ mol^-1), both enzymes possess very similar tertiary structures. The present folding studies on onconase by rapid mixing techniques in combination with fluorescence and NMR spectroscopy allow the structural assignment of the three kinetic phases observed in stopped-flow fluorescence spectroscopy. After a slow peptidyl-prolyl cis-to-trans isomerization reaction in the unfolded state, ONC folds via an on-pathway intermediate to the native state. By quenched-flow hydrogen/deuterium exchange experiments coupled with 2D NMR spectroscopy, 31 amino acid residues were identified to be involved in the structure formation of the intermediate. Twelve of these residues are identical in the RNase A sequence, which is a significantly higher percentage (39%) than the overall 28% sequence identity. Moreover, the structure of this intermediate closely resembles two of the intermediates that occur early during the refolding of RNase A. Obviously, in spite of considerable differences in their amino acid sequence the initial folding events of both proteins are comparable, guided by a limited number of conserved residues.

alpha-Amylase from mung beans (Vigna radiata) being one of the few plant alpha-amylases purified so far was studied with respect to its conformational stability by CD and fluorescence spectroscopy. The enzyme was shown to bind 3-4 Ca2+ ions, which all are important for its activity. In contrast to other alpha-amylases no inhibition was observed at high Ca2+ concentrations (100 mM). Depletion of calcium decreased the transition temperature from 87 to 48 degrees C. Kinetic stopped-flow fluorescence measurements allowed detecting two unfolding phases at > 6 M GdmCl, whereas only one phase was observed at