Local and Large-Scale Conformational Dynamics in Unfolded Proteins and IDPs. II. Effect of Temperature and Internal Friction.

Internal dynamics of proteins are essential for protein folding and function. Dynamics in unfolded proteins are of particular interest since they are the basis for many cellular processes like folding, misfolding, aggregation, and amyloid formation and also determine the properties of intrinsically disordered proteins (IDPs). It is still an open question of what governs motions in unfolded proteins and whether they encounter major energy barriers. Here we use triplet-triplet energy transfer (TTET) in unfolded homopolypeptide chains and IDPs to characterize the barriers for local and long-range loop formation. The results show that the formation of short loops encounters major energy barriers with activation energies (Ea) up to 18 kJ/mol (corrected for effects of temperature on water viscosity) with very little dependence on amino acid sequence. For poly(Gly-Ser) and polySer chains the barrier decreases with increasing loop size and reaches a limiting value of 4.6 ± 0.4 kJ/mol for long and flexible chains. This observation is in accordance with the concept of internal friction encountered by chain motions due to steric effects, which is high for local motions and decreases with increasing loop size. Comparison with the results from the viscosity dependence of loop formation shows a negative correlation between Ea and the sensitivity of the reaction to solvent viscosity (α) in accordance with the Grote-Hynes theory of memory friction. The Arrhenius pre-exponential factor (A) also decreases with increasing loop size, indicating increased entropic costs for loop formation. Long-range loop formation in the investigated sequences derived from IDPs shows increased Ea and A compared with poly(Gly-Ser) and polySer chains. This increase is exclusively due to steric effects that cause additional internal friction, whereas intramolecular hydrogen bonds, dispersion forces, and charge interactions do not affect the activation parameters.

Local and Large-Scale Conformational Dynamics in Unfolded Proteins and IDPs. I. Effect of Solvent Viscosity and Macromolecular Crowding.

Protein/solvent interactions largely influence protein dynamics, particularly motions in unfolded and intrinsically disordered proteins (IDPs). Here, we apply triplet-triplet energy transfer (TTET) to investigate the coupling of internal protein motions to solvent motions by determining the effect of solvent viscosity (η) and macromolecular crowding on the rate constants of loop formation (kc) in several unfolded polypeptide chains including IDPs. The results show that the viscosity dependence of loop formation depends on amino acid sequence, loop length, and co-solute size. Below a critical size (rc), co-solutes exert a maximum effect, indicating that under these conditions microviscosity experienced by chain motions matches macroviscosity of the solvent. rc depends on chain stiffness and reflects the length scale of the chain motions, i.e., it is related to the persistence length. Above rc, the effect of solvent viscosity decreases with increasing co-solute size. For co-solutes typically used to mimic cellular environments, a scaling of kc ∝ η-0.1 is observed, suggesting that dynamics in unfolded proteins are only marginally modulated in cells. The effect of solvent viscosity on kc in the small co-solute limit (below rc) increases with increasing chain length and chain flexibility. Formation of long and very flexible loops exhibits a kc ∝ η-1 viscosity dependence, indicating full solvent coupling. Shorter and less flexible loops show weaker solvent coupling with values as low as kc ∝ η-0.75 ± 0.02. Coupling of formation of short loops to solvent motions is very little affected by amino acid sequence, but solvent coupling of long-range loop formation is decreased by side chain sterics.

Multimodal Spectroscopic Study of Amyloid Fibril Polymorphism.

Amyloid fibrils are a large class of self-assembled protein aggregates that are formed from unstructured peptides and unfolded proteins. The fibrils are characterized by a universal ß-sheet core stabilized by hydrogen bonds, but the molecular structure of the peptide subunits exposed on the fibril surface is variable. Here we show that multimodal spectroscopy using a range of bulk- and surface-sensitive techniques provides a powerful way to dissect variations in the molecular structure of polymorphic amyloid fibrils. As a model system, we use fibrils formed by the milk protein ß-lactoglobulin, whose morphology can be tuned by varying the protein concentration during formation. We investigate the differences in the molecular structure and composition between long, straight fibrils versus short, wormlike fibrils. We show using mass spectrometry that the peptide composition of the two fibril types is similar. The overall molecular structure of the fibrils probed with various bulk-sensitive spectroscopic techniques shows a dominant contribution of the ß-sheet core but no difference in structure between straight and wormlike fibrils. However, when probing specifically the surface of the fibrils with nanometer resolution using tip-enhanced Raman spectroscopy (TERS), we find that both fibril types exhibit a heterogeneous surface structure with mainly unordered or α-helical structures and that the surface of long, straight fibrils contains markedly more ß-sheet structure than the surface of short, wormlike fibrils. This finding is consistent with previous surface-specific vibrational sum-frequency generation (VSFG) spectroscopic results ( VandenAkker et al. J. Am. Chem. Soc. , 2011 , 133 , 18030 - 18033 , DOI: 10.1021/ja206513r ). In conclusion, only advanced vibrational spectroscopic techniques sensitive to surface structure such as TERS and VSFG are able to reveal the difference in structure that underlies the distinct morphology and rigidity of different amyloid fibril polymorphs that have been observed for a large range of food and disease-related proteins.

Background-Free Fourth-Order Sum Frequency Generation Spectroscopy.

The recently developed 2D sum frequency generation spectroscopy offers new possibilities to analyze the structure and structural dynamics of interfaces in a surface-specific manner. Its implementation, however, has so far remained limited to the pump-probe geometry, with its inherent restrictions. Here we present 2D SFG experiments utilizing a novel noncollinear geometry of four incident laser pulses generating a 2D SFG response, analogous to the triangle geometry applied in bulk-sensitive 2D infrared spectroscopy. This approach allows for background-free measurements of fourth-order nonlinear signals, which is demonstrated by measuring the fourth-order material response from a GaAs (110) surface. The implementation of phase-sensitive detection and broadband excitation pulses allows for both highest possible time resolution and high spectral resolution of the pump axis of a measured 2D SFG spectrum. To reduce the noise in our spectra, we employ a referencing procedure, for which we use noncollinear pathways and individual focusing for the signal and local oscillator beams. The 2D spectra recorded from the GaAs (110) surface show nonzero responses for the real and imaginary component, pointing to contributions from resonant electronic pathways to the χ((4)) response.

Nanoscale Heterogeneity of the Molecular Structure of Individual hIAPP Amyloid Fibrils Revealed with Tip-Enhanced Raman Spectroscopy.

Type 2 diabetes mellitus is characterized by the pathological deposition of fibrillized protein, known as amyloids. It is thought that oligomers and/or amyloid fibrils formed from human islet amyloid polypeptide (hIAPP or amylin) cause cell death by membrane damage. The molecular structure of hIAPP amyloid fibrils is dominated by ß-sheet structure, as probed with conventional infrared and Raman vibrational spectroscopy. However, with these techniques it is not possible to distinguish between the core and the surface structure of the fibrils. Since the fibril surface crucially affects amyloid toxicity, it is essential to know its structure. Here the surface molecular structure and amino acid residue composition of hIAPP fibrils are specifically probed with nanoscale resolution using tip-enhanced Raman spectroscopy (TERS). The fibril surface mainly contains unordered or α-helical structures, in contrast to the ß-sheet-rich core. This experimentally validates recent models of hIAPP amyloids based on NMR measurements. Spatial mapping of the surface structure reveals a highly heterogeneous surface structure. Finally, TERS can probe fibrils formed on a lipid interface, which is more representative of amyloids in vivo.

Quantifying Surfactant Alkyl Chain Orientation and Conformational Order from Sum Frequency Generation Spectra of CH Modes at the Surfactant-Water Interface.

We combine second-order nonlinear vibrational spectroscopy and quantum-chemical calculations to quantify the molecular tilt angle and the structural variation of a decanoic acid surfactant monolayer on water. We demonstrate that there is a remarkable degree of delocalization of the vibrational modes along the backbone of the amphiphilic molecule. A simulation-based on modeled sum frequency generation (SFG) spectra offers quantitative insights into the disorder of surfactant monolayers at the water-air interface. It is shown that an average of one gauche defect in the alkyl chain suffices to give rise to the methylene stretch intensity similar in magnitude to the methyl stretch.

Model peptides uncover the role of the ß-secretase transmembrane sequence in metal ion mediated oligomerization.

The ß-secretase or ß-site amyloid precursor protein cleaving enzyme 1 (BACE1) is the enzyme responsible for the formation of amyloid-ß peptides, which have a major role in Alzheimer pathogenesis. BACE1 has a transmembrane sequence (TMS), which makes it unique among related proteases. We noticed that the BACE1 TMS contains an uncommon sulfur-rich motif. The sequence MxxxCxxxMxxxCxMxC spans the entire TMS, resembles metal ion binding motifs, and is highly conserved among homologues. We used a synthetic 31-mer model peptide comprising the TMS to study metal ion binding and oligomerization. Applying diverse biochemical and biophysical techniques, we detected dimer and trimer formation of the TMS peptide with copper ions. Replacement of the central Cys466 by Ala essentially abolished these effects. We show that the peptide undergoes a redox reaction with copper ions resulting in a disulfide bridge involving Cys466. Further, we find peptide trimerization that depends on the presence of monovalent copper ions and the sulfhydryl group of Cys466. We identified Cys466 as a key residue for metal ion chelation and to be the core of an oligomerization motif of the BACE1-TMS peptide. Our results demonstrate a novel metal ion controlled oligomerization of the BACE1 TMS, which could have an enormous therapeutic importance against Alzheimer disease.

The polyphenol EGCG inhibits amyloid formation less efficiently at phospholipid interfaces than in bulk solution.

Age-related diseases, like Alzheimer's disease and type 2 diabetes mellitus, are characterized by protein misfolding and the subsequent pathological deposition of fibrillized protein, also called amyloid. Several classes of amyloid-inhibitors have recently been tested, traditionally under bulk conditions. However, it has become apparent that amyloid fibrils and oligomers assemble and exert their cytotoxic effect at cellular membranes, rather than in bulk solution. Knowledge is therefore required of inhibitor activity specifically at the phospholipid membrane interface. Here we show, using surface-specific sum-frequency generation (SFG) spectroscopy and atomic force microscopy (AFM), that the commonly used (-)-epigallocatechin gallate (EGCG) is a much less efficient amyloid inhibitor at a phospholipid interface than in bulk solution. Moreover, EGCG is not able to disaggregate existing amyloid fibrils at a phospholipid interface, in contrast to its behavior in bulk. Our results show that interfaces significantly affect the efficiency of inhibition by EGCG inhibitors and should therefore be considered during the design and testing of amyloid inhibitors.

Time-resolved methods in biophysics. 10. Time-resolved FT-IR difference spectroscopy and the application to membrane proteins.

The introduction of time-resolved Fourier transform infrared (FT-IR) spectroscopy to biochemistry opened the possibility of monitoring the catalytic mechanism of proteins along their reaction pathways. The infrared approach is very fruitful, particularly in the application to membrane proteins where NMR and X-ray crystallography are challenged by the size and protein hydrophobicity, as well as by their limited time-resolution. Here, we summarize the principles and experimental realizations of time-resolved FT-IR spectroscopy developed in our group and compare with aspects emerging from other laboratories. Examples of applications to retinal proteins and energy transduction complexes are reviewed, which emphasize the impact of time-resolved FT-IR spectroscopy on the understanding of protein reactions on the level of single bonds.

Time-resolved flow-flash FT-IR difference spectroscopy: the kinetics of CO photodissociation from myoglobin revisited.

Fourier-transform infrared (FT-IR) difference spectroscopy has been proven to be a significant tool in biospectroscopy. In particular, the step-scan technique monitors structural and electronic changes at time resolutions down to a few nanoseconds retaining the multiplex advantage of FT-IR. For the elucidation of the functional mechanisms of proteins, this technique is currently limited to repetitive systems undergoing a rapid photocycle. To overcome this obstacle, we developed a flow-flash experiment in a miniaturised flow channel which was integrated into a step-scan FT-IR spectroscopic setup. As a proof of principle, we studied the rebinding reaction of CO to myoglobin after photodissociation. The use of microfluidics reduced the sample consumption drastically such that a typical step-scan experiment takes only a few 10 microliter [corrected] of a millimolar sample solution, making this method particularly interesting for the investigation of biological samples that are only available in small quantities. Moreover, the flow cell provides the unique opportunity to assess the reaction mechanism of proteins that cycle slowly or react irreversibly. We infer that this novel approach will help in the elucidation of molecular reactions as complex as those of vectorial ion transfer in membrane proteins. The potential application to the oxygen splitting reaction of cytochrome c oxidase is discussed.