Single-molecule fluorescence resonance energy transfer (smFRET) remains a widely utilized and powerful tool for quantifying heterogeneous interactions and conformational dynamics of biomolecules. However, traditional smFRET experiments either are limited to short observation times (typically less than 1 ms) in the case of "burst" confocal measurements or require surface immobilization which usually has a temporal resolution limited by the camera framing rate. We developed a smFRET 3D tracking microscope that is capable of observing single particles for extended periods of time with high temporal resolution. The confocal tracking microscope utilizes closed-loop feedback to follow the particle in solution by recentering it within two overlapping tetrahedral detection elements, corresponding to donor and acceptor channels. We demonstrated the microscope's multicolor tracking capability via random walk simulations and experimental tracking of 200 nm fluorescent beads in water with a range of apparent smFRET efficiency values, 0.45-0.69. We also demonstrated the microscope's capability to track and quantify double-stranded DNA undergoing intramolecular smFRET in a viscous glycerol solution. In future experiments, the smFRET 3D tracking system will be used to study protein conformational dynamics while diffusing in solution and native biological environments with high temporal resolution.

Energy barriers in kinetics are typically determined by measuring the rate as a function of temperature. We have used an alternative and complementary technique, temperature-derivative spectroscopy, to characterize folding of poly-L-glutamic acid in response to a change in pH. After the solution pH is changed photolytically at low temperature (150 K) using a photogenerating acid, the solution is warmed while monitoring the helix-coil transition by its infrared absorption. At high initial pH, the secondary structure change occurs at 180 K. If the initial pH is near 6, helix formation occurs near 190 K. The results are interpreted in terms of the pre-exponential factor and enthalpy barrier for protein folding. Because the pH change is extrinsic to the system under study, this technique enables temperature-derivative spectroscopy to be used with any phenomenon that is pH dependent and with a wide range of spectroscopic measurement techniques. (C) 2013 Elsevier B. V. All rights reserved.

Peptide side chain interactions were studied by molecular dynamics simulation using explicit solvent on a peptide with the sequence AAARAAAAEAAEAAAARA. Three different protonation states of the glutamic acid side chains were simulated for four 20 ns runs each, a total simulation time of 240 ns. Two different salt bridge geometries were observed and the preferred geometry was found to depend on Glu — Arg residue spacing. Stable charge clusters were also observed, particularly in the fully charged peptide. Salt bridges were selectively interrupted upon protonation, with concomitant changes in secondary structure. The fully charged peptide was highly helical between residues 9 and 13, although protonation increased helicity near the N-terminus. The contribution of salt bridges to helix stability therefore depends on both position and relative position of charged residues within a sequence.

The kinetics of the helix–coil transition of poly-l-glutamate were measured in the range of 40 ns to 10 s using a laser-induced pH-jump coupled with time-resolved infrared spectroscopy. Folding of the polypeptide in D2O was initiated by photolyzing o-nitrobenzaldehyde, which releases a deuteron, creating a rapid decrease in pD. Side-chain deuteration and conformational changes were monitored independently by varying the IR probe wavelength. The kinetics of the peptide conformational changes observed in the amide I region depended on the initial fraction of helical residues. With essentially no initial helix, amide I absorption changes were indistinguishable from those of instrument response, leading to the conclusion that helix initiation occurs in less than 40 ns. When the initial helix fraction is 0.13, the folding lifetime is lengthened to 625 ns, as predicted by helix–coil theory. We also observe evidence for a kinetically-trapped, nonproductive intermediate formed as the result of rapid deuteration of the unfolded state.

FTIR difference spectroscopy is used to reveal changes in the internal structure and amino acid protonation states of bovine cytochrome c oxidase (CcO) that occur upon photolysis of the CO adduct of the two-electron reduced (mixed valence, MV) and four-electron reduced (fully reduced, FR) forms of the enzyme. FTIR difference spectra were obtained in D2O (pH 6–9.3) between the MV-CO adduct (heme a3 and CuB reduced; heme a and CuA oxidized) and a photostationary state in which the MV-CO enzyme is photodissociated under constant illumination. In the photostationary state, part of the enzyme population has heme a3 oxidized and heme a reduced. In MV-CO, the frequency of the stretch mode of CO bound to ferrous heme a3 decreases from 1965.3 cm−1 at pH* ≤7 to 1963.7 cm−1 at pH* 9.3. In the CO adduct of the fully reduced enzyme (FR-CO), the CO stretching frequency is observed at 1963.46±0.05 cm−1, independent of pH. This indicates that in MV-CO there is a group proximal to heme a that deprotonates with a pKa of about 8.3, but that remains protonated over the entire pH* range 6–9.3 in FR-CO. The pKa of this group is therefore strongly coupled to the redox state of heme a. Following photodissociation of CO from heme a3 in MV oxidases, the extent of electron transfer from heme a3 to heme a shows a pH-dependent phase between pH 7 and 9, and a pH-independent phase at all pH's. The FTIR difference spectrum resulting from photolysis of MV-CO exhibits vibrational features of the protein backbone and side chains associated with (1) the loss of CO by the a3 heme in the absence of electron transfer, (2) the pH-independent phase of the electron transfer, and (3) the pH-dependent phase of the electron transfer. Many infrared features change intensity or frequency during both electron transfer phases and thus appear as positive or negative features in the difference spectra. In particular, a negative band at 1735 cm−1 and a positive band at 1412 cm−1 are consistent with the deprotonation of the acidic residue E242. Positive features at 1552 and 1661 cm−1 are due to amide backbone modes. Other positive and negative features between 1600 and 1700 cm−1 are consistent with redox-induced shifts in heme formyl vibrations, and the redox-linked protonation of an arginine residue, accompanying electron transfer from heme a3 to heme a. An arginine could be the residue responsible for the pH-dependent shift in the carbonyl frequency of MV-CO. Specific possibilities as to the functional significance of these observations are discussed.

The thermal unfolding of a series of 6-, 10-, and 14-mer cyclic beta-hairpin peptides was studied to gain insight into the mechanism of formation of this important secondary structure. The thermodynamics of the transition were characterized using temperature dependent Fourier transform infrared spectroscopy. Thermodynamic data were analyzed using a two-state model which indicates increasing cooperativity along the series. The relaxation kinetics of the peptides in response to a laser induced temperature jump were probed using time-resolved infrared spectroscopy. Single exponential relaxation kinetics were observed and fit with a two-state model. The folding rate determined for these cyclic peptides is accelerated by some two orders of magnitude over the rate of a linear peptide that forms a beta-hairpin. This observation supports the argument that the rate limiting step in the linear system is either stabilization of compact collapsed structures or rearrangement of collapsed structures over a barrier to achieve the native interstrand registry. Small activation energies for folding of these peptides obtained from an Arrhenius analysis of the rates imply a primarily entropic barrier, hence an organized transition state having specific stabilizing interactions.

The helix is a common secondary structural motif found in proteins, and the mechanism of helix-coil interconversion is key to understanding the protein-folding problem. We report the observation of the fast kinetics (nanosecond to millisecond) of helix melting in a small 21-residue alanine-based peptide. The unfolding reaction is initiated using a laser-induced temperature jump and probed using time-resolved infrared spectroscopy. The model peptide exhibits fast unfolding kinetics with a time constant of 160 +/- 60 ns at 28 degrees C in response to a laser-induced temperature jump of 18 degrees C which is completed within 20 ns. Using the unfolding time and the measured helix-coil equilibrium constant of the model peptide, a folding rate constant of approximately 6 x 10(7) s-1 (t1/2 = 16 ns) can be inferred for the helix formation reaction at 28 degrees C. These results demonstrate that secondary structure formation is fast enough to be a key event at early times in the protein-folding process and that helices are capable of forming before long range tertiary contacts are made.

The primary objective of this work was to develop a molecular understanding of how proteins achieve their native three-dimensional (folded) structures. This requires the identification and characterization of intermediates in the protein folding process on all relevant timescales, from picoseconds to seconds. The short timescale events in protein folding have been entirely unknown. Prior to this work, state-of-the-art experimental approaches were limited to milliseconds or longer, when much of the folding process is already over. The gap between theory and experiment is enormous: current theoretical and computational methods cannot realistically model folding processes with lifetimes longer than one nanosecond. This unique approach to employ laser pump-probe techniques that combine novel methods of laser flash photolysis with time-resolved vibrational spectroscopic probes of protein transients. In this scheme, a short (picosecond to nanosecond) laser photolysis pulse was used to produce an instantaneous pH or temperature jump, thereby initiating a protein folding or unfolding reaction. Structure-specific, time-resolved vibrational probes were then used to identify and characterize protein folding intermediates.

We have used picosecond infrared (IR) transient absorption spectroscopy in the amide I band tp probe the dynamics of protein motion of myoglobin (Mb) following the photolysis of carbon monoxide. The rise time of the deoxy protein conformation is shown to be about 8 ps. The spectrum of amide I changes was also measured at 50 ps after photolysis and found to be similar to static IR difference spectra and to time-resolved IR spectra taken at times longer than 100 ns. By comparing the results obtained here with other picosecond results on photolysis of CO from Mb, we conclude that the majority of changes seen in the amide I spectra are due to global motion on the proximal side of the heme. The time scale for amide I changes are compared to the results of molecular dynamics calculations.

A critical feature of the biological function of heme proteins is the direct coupling of protein motion to the process of binding exogenous ligands to the heme. In carbonmonoxymyoglobin (MbCO), a substantial, specific conformational relaxation is associated with the transition from the ligated to the unligated form of the protein. The analogous tertiary structural changes of the monomer heme subunits of hemoglobin ultimately lead to the R→T quaternary structural transition, the allosteric control mechanism of O 2 binding efficiency [1]. We have studied these processes on the earliest timescales, using picosecond, time-resolved infrared (TRIR) spectroscopy. It has long been known that infrared spectra in the amide region are sensitive to protein secondary conformation [2]. Recent advances in equipment and techniques have permitted researchers to quantitatively predict secondary structures from infrared spectra [3,4], particularly in the amide I region [4]. Therefore, it is now possible to study protein motion in time-resolved experiments on dynamics and function. The ligation reactions of small molecules such as CO with the heme site of Mb exemplify the mechanisms available to O 2 . CO is an ideal candidate for initial time-resolved IR experiments in the amide I region because it is easily photolyzed, little geminate recombination [5], and the structure of both MbCO and unligated Mb have been studied by crystallographic methods [6]. TRIR has already been applied to the stretching vibrations of the bound and free CO ligand [7,8]; dynamics of the protein, however, have yet to be probed by TRIR spectroscopy of the protein vibrations. Here we report results on the motions of the protein in response to ligation reactions, probed in the amide I region centered about 1650 cm -1 .

A central question in Mb ligand binding is the dynamics and energetics of motion of the protein in response to ligation. Time-resolved infrared (TRIR) spectroscopy is uniquely suited as a probe for these processes, particularly the behavior of the protein, which generally is not observable by other spectroscopies [1].