Sizing up single-molecule enzymatic conformational dynamics.

Published on 04 December 2013. Downloaded by TECHNISCHE UNIVERSITEIT EINDHOVEN on 17/06/2014 02:36:02.

Chem Soc Rev View Article Online

REVIEW ARTICLE

Cite this: Chem. Soc. Rev., 2014, 43, 1118

View Journal | View Issue

Sizing up single-molecule enzymatic conformational dynamics H. Peter Lu Enzymatic reactions and related protein conformational dynamics are complex and inhomogeneous, playing crucial roles in biological functions. The relationship between protein conformational dynamics and enzymatic reactions has been a fundamental focus in modern enzymology. It is extremely difficult to characterize and analyze such complex dynamics in an ensemble-averaged measurement, especially when the enzymes are associated with multiple-step, multiple-conformation complex chemical interactions and transformations. Beyond the conventional ensemble-averaged studies, real-time singlemolecule approaches have been demonstrated to be powerful in dissecting the complex enzymatic reaction dynamics and related conformational dynamics. Single-molecule enzymology has come a long way since the early demonstrations of the single-molecule spectroscopy studies of enzymatic dynamics about two decades ago. The rapid development of this fundamental protein dynamics field is hand-inhand with the new development of single-molecule imaging and spectroscopic technology and methodology,

theoretical

model

analyses,

and

correlations

with

biological

preparation

and

characterization of the enzyme protein systems. The complex enzymatic reactions can now be studied one molecule at a time under physiological conditions. Most exciting developments include active manipulation of enzymatic conformational changes and energy landscape to regulate and manipulate Received 9th June 2013

the enzymatic reactivity and associated conformational dynamics, and the new advancements have

DOI: 10.1039/c3cs60191a

established a new stage for studying complex protein dynamics beyond by simply observing but by

www.rsc.org/csr

actively manipulating and observing the enzymatic dynamics at the single-molecule sensitivity temporally and spatially.

Bowling Green State University, Center for Photochemical Sciences, Department of Chemistry, Bowling Green, OH 43403, USA. E-mail: [email protected]

H. Peter Lu received his Ph.D. degree (1991) from Columbia University. He undertook postdoctoral research at the Northwestern University (1991–1995), and subsequently at the Pacific Northwest National Laboratory (1995–1996). He worked as a Senior Research Scientist and then a Chief Scientist at Pacific Northwest National Laboratory (1996–2006). He has been an Ohio Eminent Scholar and a Full Professor of Chemistry at BowlH. Peter Lu ing Green State University in Ohio since 2006. His research focuses on single-molecule spectroscopy studies of molecular kinetics and dynamics in condensed phase and at interfaces, involving studies of protein conformational dynamics and function rate processes.

1118 | Chem. Soc. Rev., 2014, 43, 1118--1143

1. Introduction In a living cell, enzyme proteins participate in many critical biological processes. Enzymes can change the biological reaction pathways and accelerate the reaction rate by millions of times. Many critical biological functions cannot be activated and completed without enzymes. Enzymology has been a core research area in the modern life science; accordingly, enzymatic reaction dynamics and mechanisms have been a central focus of biophysics and chemical physics. Enzymatic reactions involving complex dynamics of substrate–enzyme binding, substrate–enzyme complex formation, catalytic reaction, and product release typically show temporal and spatial disorders.1–43 In recent years, single-molecule spectroscopy has been demonstrated to be a powerful approach for characterizing such complex dynamics by probing one enzyme molecule at a time, capable of recording single-molecule time trajectories of specific conformational changes and catalytic turnover cycles.1–77 The active-site conformational dynamics and mechanism of the enzymes, especially, the conformational dynamics in the interactions of the enzyme and the substrate as well as the enzymatic turnovers, are under intensive studies. The relationship

This journal is © The Royal Society of Chemistry 2014

View Article Online


Review Article

between enzyme conformational dynamics and enzymatic reaction turnovers has been a fundamental focus in modern enzymology.1–39,48,52,62–73,75–129 Conventional enzymatic kinetics and assays have been extensively applied for studying enzymatic reactions by ensembleaveraged measurements. Real-time enzyme protein motions are not measurable in such ensemble-averaged measurements as these conformational motions are stochastic and multiple-step in nature. Obviously, enzyme molecules do not act alone and typically engage in complex and dynamic interactions with other molecules or local biological components of their environments, for example, in living cells. Enzyme protein conformations are highly dynamic rather than static, and they play a critical role in enzymatic reaction activity, such as forming enzyme–substrate complexes and release products. Furthermore, most of the critical active-site conformational changes and ordered–disordered protein structure transitions2–6,8–9,11,12,14–16,18,20,24,31,36,40–43,58,62,63,65–77,130–142 are rather complex and flexible both temporally and spatially. In recent years, as more and more protein static structures have been resolved and analyzed, it has been increasingly recognized that only the static structure knowledge alone is not sufficient to understand real-time protein functions.1–39,44–47,49–53,55–61,85,86,105 Although static structure analyses from ensemble-averaged measurements at equilibrium are critical for understanding protein structure and functions, it is definitely critical to probe the non-equilibrium dynamics of the conformations under biological activities. In recent years, a significant number of excellent and comprehensive overall review articles on the current field of single-molecule spectroscopy and even single-molecule enzymology3,10,13,17,24,25,27,36,58,74,105,110,111 have been published, including the articles in this particular special series of review articles on single-molecule spectroscopy. Nevertheless, instead of presenting another article to review and comment on the current developments and status of the field of the singlemolecule enzymatic conformational dynamics, we will focus our discussions on the primary concepts of enzymatic activesite conformational dynamics by using some of our recent published works as examples. 1.1 Static and dynamic protein structure to function relationship There is a real need for complementary knowledge about both the protein static structures and their dynamics. Both static and dynamic structure analyses using techniques such as X-ray crystallography and NMR spectroscopy are powerful and informative in providing the critical understanding of the protein structures and, more often, the ensemble-averaged structures under specific equilibrium conditions.20,21,26,115–120,122,143–145 These structural specifications provide the crucial foundation for many studies of the protein conformational dynamics, at both ensembleaveraged level and single-molecule level.20,21,26,115–120,122,143–145 There are significant works and advancements published in the literature, which will not be the focus of our review article. Nevertheless, the complete protein function characterization is not complete only if the structure dynamics is also analyzed under real activity conditions. Overall, there are a few primary


Chem Soc Rev

challenges and needs for conventional protein structure–function analyses: (1) the ensemble-average measurements average out the conformational fluctuation dynamics that is often critical for enzymatic reaction dynamics; (2) protein structure rate processes can be controlled by thermodynamics or by kinetics or by both, and it is often that a bifurcation pathway can be kinetically controlled and hard to be detected by static structure analysis or averaging kinetic analyses, due to the fact that the pathway determining intermediate states are relatively short-lived compared to the majority background; (3) limited time resolution and structure volume resolutions; and (4) significant deviation of the measurement condition from the real-time enzymatic reaction conditions, including the concentrations of enzyme proteins and substrates, temperature, solvent buffer conditions, etc. Single-molecule spectroscopy has been demonstrated as a powerful complementary approach for characterizing dynamic conformations and understanding correlated protein functions in real time, typically, by detecting single-molecule time trajectories. Single-molecule spectroscopy and the related combined single-molecule approaches enable characterization of temporal, spatial, and energetic properties of protein conformations that often play critical roles in complex interactions amongst protein molecules, enzyme–substrate interactions under enzymatic reaction conditions, and ion channel processes in membranes.1–39,44–47,49–53,55–61,85,86,146 Single-molecule studies have been providing mechanistic understanding of some of the most important biological processes, including receptor sensing, cell redox respirations, cell signaling, and enzymatic reactions. Several excellent articles10,13,17,20,29–31,33,34,36,37,47,51,54,55,105 have provided extensive reviews of single-molecule spectroscopy and its applications.1–39,44–47,49–53,55–61,85,86,105 For example, single-molecule fluorescence photon-stamping time trajectory detection3,9,29,36,44–47,49–53,55–61,85,86,93,101,107–109 and statistical analysis have revealed static and dynamic inhomogeneities in the dynamics of a number of protein systems.2–6,9,11,12,15,24,25,27,29–31,38,40–47,49–53,55–61,85,86 Static inhomogeneity describes inhomogeneous molecular dynamics from molecule to molecule, whereas dynamic inhomogeneity originates in the temporal fluctuations of the kinetic rates of individual molecules, which is beyond the scope of conventional kinetics.2–6,8,9,11,12,15,24,25,27,29,38,40–47,49–53,55–61,85,86 It is nearly impossible for an ensemble-averaged measurement to selectively identify and characterize both static and dynamic inhomogeneities. 1.2 Single molecule approaches – studying one enzyme molecule at a time to dissect complex and inhomogeneous biological dynamics Studying one enzyme molecule at a time in single-molecule spectroscopy has significant advantages over conventional ensemble-averaged measurements: (1) Enzymatic conformational fluctuation and dynamics are intrinsically stochastic, and an ensemble-averaged measurement is only able to yield a mean value of the fluctuating properties. However, a singlemolecule measurement directly obtains the time trajectories of physical-property changes in real time to follow single protein

Chem. Soc. Rev., 2014, 43, 1118--1143 | 1119

View Article Online


Chem Soc Rev

actions. (2) Enzymatic reactions typically involve multiple step mechanisms, and even for the simplest type, the reaction follows the Michaelis–Menten mechanism which includes at least three steps.3,24–26,29–31,35,39,44,54,55,57 It is nearly impossible to perform step-by-step synchronization of complex and multiple-step biological interactions and transformations at an ensemble level. Nevertheless, single-molecule spectroscopic measurements do not need synchronization of an ensemble of molecules; therefore, the non-synchronizable, multiple-step, and multiple-state enzymatic reaction processes can be specifically probed and analyzed. (3) Single-molecule approaches are able to remove and identify temporal and spatial inhomogeneities of the enzyme–substrate complex interactions and enzymatic reaction processes; therefore, enzymatic conformational cooperativity, hysteresis, and non-Markovian conformational and catalytic dynamics can be identified and characterized in detail. (4) Single-molecule approaches are able to identify and differentiate between static inhomogeneity and dynamic inhomogeneity, whereas it is nearly impossible for an ensemble-averaged measurement to selectively identify and characterize static and dynamic inhomogeneities. (5) Biological systems, such as enzymatic reactions and cell signaling, often intrinsically involve in competitive multiple-pathway dynamics of both parallel and consecutive pathways under either kinetic or thermodynamic control. By selectively recording single-molecule time trajectories, single-molecule spectroscopy is capable of identifying transient kinetically controlled intermediate states and characterizing the complex mechanisms.

2. The specific techniques and methods of single-molecule spectroscopy and imaging Here we focus our discussion on a few specific single-molecule spectroscopic approaches that are directly related to the studies on enzyme conformational and catalytic dynamics. These approaches are typically able to record stochastic trajectories of a single-molecule property in real time. These approaches characterize the inhomogeneities of complex biological systems by studying one molecule or molecular complex at a time. These trajectories reveal intrinsic single-molecule conformational dynamics and local environmental change dynamics through calculation of first-order and higher-order autocorrelation functions as well as cross-correlation functions from the trajectories.2–6,8–9,11,12,14,16,18,29,36,44,54,55,57,108,109 2.1

Single-molecule photon stamping spectroscopy

Single-molecule photon stamping spectroscopy has been developed, applied and demonstrated recently in a number of laboratories,3,9,29,31,36,44–47,49–53,55–65,85,86,93,101,107–109,136,138 each with some home-developed special features. Generally speaking, for each detected photon, we are able to record two key parameters: the chronic detection time, i.e., photon-detection recording real time, of the photon and the specific photon emission time delay from the molecule’s excited state. For binning in chronic time

1120 | Chem. Soc. Rev., 2014, 43, 1118--1143

Review Article

Fig. 1 Schematic diagram of a combined AFM-photon stamping singlemolecule spectroscopy set up. Various CW and/or pulse lasers are used to provide imaging and excitation light focused on the sample through an inverted microscope with various configurations of illuminations, including wide-field, confocal, and total internal reflection illuminations. The optical focus and the AFM scanning probe tip are positioned in an over-and-under configuration at the microscopic focal point. The single-molecule protein sample can be studied either simultaneously or consecutively by topographic, spectroscopic, and combined manipulation and detection. The detailed explanation of the set up can be found in our publications.44–47,49–53,55–61,85,86

from the single-molecule photon stamping trajectories, we have the conventional single-molecule intensity trajectories, and from each photon time delay, we have attempted to measure the excited state lifetime once. An analogy of the photon stamping detection recording both detection time and time delay of photon emission from the photon excitation can be like receiving a mail with two stamps on the envelope – one marking the time the mail is sent out from the sending post office and other the time the mail arrived at the receiving post office. In our experiments (Fig. 1 and 2), photon stamping can be done in two ways: (1) A continuous-wave (CW) laser is used for measuring the time trajectory of the single-molecule fluorescence intensity fluctuation by detecting each photon and ‘‘stamping’’ (recording) its arrival time. This technique has pushed the single-molecule fluorescence intensity detection to the detection limit so that the time sequence of every detected photon is recorded, and subsequent off-line binning (taking an average of a certain number of channels consecutively) provides the highest time resolution. The time resolution is only limited by the photon flux, excitation saturation, photobleaching, and detector counting rate. (2) A pulsed-laser based approach is used in which each emitted photon is detected and its delay time stamped, Dt, from the pump laser pulse. The delay time is the duration from the time of excitation of the single molecule to the time of emission by the same molecule, which is a single event measurement of the fluorescence lifetime of this molecule. As a Poisson process, the distribution of the delay time gives an exponential decay, and the statistical mean of the distribution reports the value of the single-molecule fluorescence lifetime.


View Article Online


Review Article

Chem Soc Rev

Since EFRET is sensitive to changes in the distance between the donor and the acceptor, the energy transfer efficiency can also be determined by probing the fluorescence intensity of the donor (ID) and the acceptor (IA):97,102 EFRET = IA/[IA + ID]

(2)

Therefore, by labeling the donor and acceptor dye probes in the specific positions of a protein and measuring single-molecule EFRET(t) B t trajectories, the time-resolved single-molecule conformational dynamics can be probed.2–6,9,11,12,15,24,25,27,29–31,38, 44–53,55–61,77,79,83–86,90,102,104,106

Fig. 2 Photon stamping concept and definition of the parameters: (A) scheme of a train of laser excitation pulses and detected emission photons; (B) scheme of the time-stamped photon sequence. The delay time tp is the time delay between the photoexcitation event and the photon emission; the real time tp is the chronic time of detecting emission photons; for each detected photon, both tp and tp are simultaneously recorded. (C) A typical experimental photon stamping trajectory. Each data point represents a detected photon event with two physical parameters recorded, the delay time tp and the real time tp of detection. The sorted distribution along the y-axis of tp gives the conventional time correlated photon counting lifetime decay curve, and the binning along the x-axis of tp gives the conventional single-molecule photon counting intensity trajectory.

2.2 Fluorescence intensity and lifetime based single-molecule FRET spectroscopy Fluorescence resonance energy transfer (FRET) is the energy transfer from the donor to the acceptor via non-radiative transition dipole– dipole interactions. The energy transfer efficiency (EFRET) depends on the distance between the donor and the acceptor:97,102,105,110–114 EFRET = 1/[1 + (r/R0)6]

(1)

where r is the separation distance between two fluorescent dyes, the FRET donor and acceptor, which typically is in the range of ¨rster radius is the separation distance 20–80 Å. R0 being the Fo when EFRET equals 50% and is dependent on the dipole orientation of two dyes, the refractive index of the medium, the spectral overlap integral between donor emission and acceptor absorbance, and the quantum yield of the donor.97,102 In this session, we will focus our discussion on lifetime based single-molecule FRET spectroscopy, after a brief description of the widely discussed intensity-based single-molecule FRET spectroscopy.


In recent years, there are extensive demonstrations and applications of using the fluorescence intensity based singlemolecule FRET measurements for probing the biomolecular conformational changes and motions. Molecule conformational changes, protein motions, and protein–DNA and protein–protein interactions can be monitored by exciting the donor and monitoring either the donor’s or the acceptor’s intensity trajectories, or both simultaneously.2–6,9,11,12,24,25,27,29,36,49–54,56–60,105,108 Site-selectively labeling DNA with a donor and an acceptor on either side of the damaged site enables one to observe the binding-induced conformational changes by monitoring donor and/or acceptor fluorescence intensity and lifetime fluctuations due to singlepair FRET. Single-molecule spFRET studies of enzymatic reactions, RNA and ribozyme conformational changes, and a few other important biological processes have been reported in recent years.2–6,9,11,12,24,25,27,29,36,48–54,56–60,83,90,106 Due to photobleaching, probe molecule diffusion motions, and spectral fluctuations of single molecules, spFRET is a reliable and effective method for probing distance change dynamics but less reliable for measuring exact distances. Nevertheless, it is often a challenge of obtaining high signal-to-noise ratio for using the intensity-based single-molecule FRET to measure conformational changes of the enzyme, especially the active-site conformational dynamics. This is due to a simple fact: the protein conformational changes typically show only small fractional changes in FRET efficiency since the change in the donor– acceptor distance is often only 1–2 nm for the enzyme activesite conformational changes.5,6,9,10,13,34,36,59 Specifically, the single-molecule FRET measurements often encounter much smaller changes in FRET efficiency in probing protein conformational dynamics9,24,29,45,46,50 than, for example, in probing DNA/ RNA conformational dynamics.25,36,105 Therefore, experimentally, a typical FRET intensity-based efficiency measurement is particularly susceptible to measurement error and background noise when the experimental parameters are variable and time dependent during the measurements. FRET time trajectories associated with single-molecule conformational dynamics can be recorded by measuring the donor’s lifetime fluctuations, since the energy transfer from the donor to the acceptor via transition dipole–dipole interactions decreases the donor’s fluorescence lifetime. This is the central concept of the fluorescence lifetime based singlemolecule FRET spectroscopy. The donor’s fluorescence lifetime decreases with increasing FRET efficiency and vice versa, according to the equation EFRET = 1 tDA/tD, where tDA and

Chem. Soc. Rev., 2014, 43, 1118--1143 | 1121

View Article Online


Chem Soc Rev

tD are the donor’s lifetime in the presence (tDA) and in the absence (tD) of the acceptor, i.e., with and without the FRET interactions, respectively. There are significant technical developments on single-molecule photon stamping spectroscopy and its application in lifetime based FRET measurements.45,46,83,97,105 Compared with single-molecule fluorescence intensity-based FRET measurements, single-molecule lifetimebased FRET measurements are independent of fluorescence intensity, which has the advantage in terms of eliminating the analysis background noise from acceptor fluorescence detection leak through noises, excitation light intensity noises, or light scattering noises due to local environmental factors, for example, avoid the photon counting background perturbation in an AFM-tip correlated single-molecule FRET measurement.45,77,79,83,84,97,102,104,109 Furthermore, the lifetimebased FRET also supports simultaneous single-molecule fluorescence anisotropy, which will be discussed below. Here we show an example of how a lifetime based singlemolecule FRET measurement is carried out. Fig. 3 shows a typical single-molecule lifetime trajectory collected from a Cy3–Cy5 labeled kinase molecule under enzymatic reaction conditions with substrates, which illustrates the basis of single molecule donor’s lifetime trajectory tDA(t) B t measurements. In this measurement, we record each fluorescence photon’s real arrival time tp and each fluorescence photon’s delay time tp related to laser pulse excitation (Fig. 3). Fig. 3A and B show donor–acceptor two channel images and a portion of singlemolecule photon stamping raw data from the donor channel in a period of 0.8 seconds, respectively. For each detected photon, we recorded two parameters: a chronic arrival time (t) and a delay time related to femtosecond laser pulse excitation (tp) (Fig. 1–3). The chronic arrival times of the fluorescence photons contain information about the photon flux so that we can count and bin the photons at a given time scale, for example, of 10 ms binning time to record a typical fluorescence intensity trajectory. The lifetime (tDA) in each 10 ms bin (Fig. 3B) can be obtained by fitting the histogram of the delay time of all the photons in 10 ms bins with an exponential function or by calculating the mean of the delay time of all of the photons in 10 ms bins. We typically treat the photon counting distribution in each bin as a Poisson distribution that gives the means of each distribution as fluorescence lifetime, tDA.45,46,83,97,98,105,106 In Fig. 3C, we show typical histograms of the delay time of the fluorescence photons in 10 ms bins under different FRET states. The decrease in fluorescence lifetime, comparing left and right panels, indicates an increase of the FRET efficiency, most likely resulting from the decrease of the donor–acceptor separation distance (insets in Fig. 3B). After calculating all the delay times in each 10 ms bin from the original photon stamping data (Fig. 3B), and connecting the lifetime data resolved in all of the time bins, we obtain the lifetime trajectory of the donor (tDA B t, shown in Fig. 3D). The lifetime trajectory of the donor in FRET reports the EFRET fluctuation associated with the conformational fluctuation of the enzyme.45,46 Therefore, by analyzing the lifetime trajectory of the donor, we are able to detect conformational dynamics, probing real-time donor–acceptor distance changes.

1122 | Chem. Soc. Rev., 2014, 43, 1118--1143

Review Article

Fig. 3 Single-molecule photon-stamping measurement and data analysis study of lifetime-based FRET fluctuations; all the data are collected from a Cy3–Cy5 labeled single HPPK molecule under the enzymatic reaction condition with 100 mM ATP and 100 mM HP. (A) Single-molecule FRET donor– acceptor (left and right panels) two-channel imaging by photon stamping spectroscopy. (B) An example of the single-molecule photon-stamping raw data from the donor channel in a 0.8 seconds period (5.8–6.6 s). Each data point corresponds to a detected photon plotted by the delay time (tp) vs. its chronic arrival time (tp). (C) Histograms of the delay times of the photons in a 10 ms period from the trajectory shown in (A). The left panel is the histogram of delay times in 10 ms (6.08–6.09 s), corresponding to the low energy transfer efficiency from the donor to the acceptor. The right panel is the histogram of delay times in 10 ms (6.30–6.31 s), corresponding to the high energy transfer efficiency. (D) Lifetime trajectory of the donor (tDA) calculated from the trajectory in (A) with 10 ms binning. The arrows show the positions (6.08–6.09 s) and (6.30–6.31 s) of the lifetime trajectory. We used femtosecond pulse laser excitation and a home-built two-channel single-molecule FRET lifetime microscope, and the detailed description of our measurements and instrumentation has been reported previously.9,11,12,44–47,49–53,55–60 (Adapted with permission from ref. 45. Copyright 2013 American Chemical Society.)

2.3

Single-molecule static and transient anisotropy

The rates of rotational motion of molecules are dependent upon molecular hydrodynamic shapes and hydrodynamic volume that can be changed by conformational changes, molecular interactions, and complex formation and dissociation. Single-molecule rotational motions can be probed by using parallel and perpendicular-polarized two-channel detection (Fig. 4). This measurement can be conducted using either a CW laser or a pulse laser and, therefore, the time-resolution is sub-microseconds to seconds, which is suitable to study molecular rotational jumps and slow molecular conformational motions.36,46,48,80,83,147 To probe faster conformational motions, single-molecule time-resolved fluorescence anisotropy has been developed and applied by using the pulse laser excitations, and this technique has significantly


View Article Online


Review Article

Chem Soc Rev

labeled calmodulin (CaM-Texas Red). For CaM-Texas Red, it was difficult to resolve the protein motion dynamics due to rapid emission depolarization, which indicated rotational mobility of the fluorescent label relative to the host protein. In contrast, the parallel and perpendicular fluorescence decay component of CaMFlAsH yielded a rotational lifetime of B7.2 ns characteristic of calmodulin (15 kDa). This demonstrates that FlAsH is rigidly attached to the host protein, thus allowing for the unambiguous probing of the calmodulin orientational dynamics without convolution of the probe-dye motions.59 2.4 Protein conformation manipulation by AFM-FRET ultramicroscopy

Fig. 4 (A) A cartoon representation of the T4 lysozyme protein tethered to a hydrocarbon-modified glass surface through a bi-function linker. The enzyme can freely move in buffer solution and its enzymatic activity is not perturbed by the surface. (B) The parallel and perpendicular fluorescence decay and single exponential fits of a single T4 lysozyme/Alexa 488 molecule covalently linked to the surface. The decay curves are integrated from all detected photons of parallel and perpendicular channels before photobleaching. (C) Using singlemolecule photon stamping spectroscopy to study the fluorescence intensity, lifetime, and anisotropy of a single molecule simultaneously by recording the arrival time and delay time of each fluorescence photon. The top-left plot is an example of the raw data of the photon time-stamping TCSPC of a detector channel. Each dot corresponds to a photon detected, plotted by its arrival time (t) and delay time (Dt) (raw output from TAC in reverse timing). The fluorescence intensity trajectory (not shown) can be calculated from the histogram of arrival time (t) with a given time-bin resolution. The molecule was photo-bleached at 8.71 seconds. The nanosecond fluorescence decay curves (top-right plot) are the histograms of the delay time of the fluorescence photons (t o 8.1 s) and background photons (t > 8.1 s).

enhanced the single-molecule anisotropy measurement of the protein rotational motions and conformational changes from sub-milliseconds46,50,147 to nanoseconds (Fig. 4). The capability of measuring the nanosecond anisotropy of a site-specifically labeled dye molecule has been applied to probe the conformational dynamics of the protein.24,36,46,50,147 Recently, a new dye-labeling technique has provided covalently attached dye-probe molecules, so-called FlAsH labeling.59,148,149 The rotational motions of these molecules are locked on protein matrixes that contain a tetra-cysteine motif, which is genetically inserted within a b-helix region near the amino terminus. The tetra-cysteine motif then reacts with FlAsH, a fluorescein derivative with two As(III) substituents, which fluoresces only after the arsenics bind to cysteine thiols.59,148,149 The tight arsenicbinding structure ensures that the fluorescent label has no rotational freedom relative to the protein matrix so that the protein motions can be probed unambiguously without convolution of fast dye molecule motions. For example, we have tested FlAsH as an effective fluorescent label for probing protein motion dynamics. We have demonstrated a single-molecule nanosecond anisotropy measurement of FlAsH labeled calmodulin (CaM-FlAsH) and Texas Red


As a new frontier of the single-molecule approaches to study enzymatic conformational dynamics and catalytic dynamics, correlated multiple channel analytical single-molecule microscopes have been developed and demonstrated in recent years.45,46,74,103,114,121,123,124,150–154 The clear advantage of such new approaches is that the analyses of the protein dynamics now can be done under active manipulations specifically for the protein coordinates, domains, and active sites, which have been proved to be highly powerful and unique in resolving the complex catalytic dynamics and associated protein conformational dynamics. In Fig. 5 and 6, we show the concept of the experimental setup of AFM-FRET nanoscopy. Generally, there are high technical challenges for AFM-FRET single-molecule nanoscopy to operate: (1) lining up the optical focal point and the AFM tip for a typical operation of our AFM-FRET nanoscopy and (2) eliminating the micro-mirror reflection effect of the AFM tip on the photon detection of the single-molecule imaging. Typically, after aligning the AFM tip with the laser beam focus in an over-and-under co-axial configuration, we first record an optical image (10 mm 10 mm) by raster scanning the closedloop 2D electropiezo-scanning stage with the sample over the laser focus at a scanning speed of 4 ms per pixel. We had developed two imaging approaches to make sure that the co-axial alignment is within the range of a few nanometers in diameter: (1) lifetime correlated AFM imaging and (2) AFM matrix mapping. We have developed and demonstrated a lifetime correlated AFM imaging approach capable of obtaining topographic and spectroscopic images at nanoscale spatial resolutions, using a metal coated AFM tip positioned in an over-and-under configuration with the microscopic imaging laser excitation light. The essential concept of this approach is that the metal tip at the nanometer scale only perturbs the targeted nanostructure or molecule in the range of nanometers. For example, if a simultaneous AFM topographic and fluorescence lifetime imaging scan is conducted within the laser focus spot of B300 nm diameter, the molecule fluorescence lifetime changes only when the metal AFM tip is right on top of the targeted nanostructure or molecule within the nanometer scale, and the lifetime recovers as long as the tip moves away from the molecule by a few nanometers. In Fig. 5B, we show the same spatial resolution of fluorescence-lifetime AFM microscopic imaging of an B40 nm polymer bead doped with fluorescent molecules, and the same spatial resolution is

Chem. Soc. Rev., 2014, 43, 1118--1143 | 1123

View Article Online


Chem Soc Rev

Fig. 5 Correlated single-molecule AFM-FRET and AFM-photon stamping spectroscopic microscopy station. (A) Field emission scanning electron micrograph of the Au-coated AFM tip. A globular Au structure at the apex of the tip is evident. (B) The two images are from different signals during the same image scan. (Upper) The topography of the nanosphere from AFM. (Lower) The fluorescence lifetime of the nanosphere as an effect of tip scanning. The circle and the rectangle are the areas used to fit the fluorescence decay traces with and without tip effect, respectively. It is evident that the metal tip enhanced lifetime imaging can reach the same spatial resolution of a typical AFM imaging measurement. (C) A schematic configuration of the correlated single-molecule AFM-FRET microscope. The AFM tip and optical microscopic imaging are in an over-and-under configuration. (D) A typical imaging of making the alignment of both the AFM tip from the top and the focused laser illumination spot from the bottom in a coaxial position. The green line shows a scanning trace of the AFM tip.

reached for both topographic and spectroscopic images. This high spatial resolution imaging approach is capable of supporting optical imaging and analysis of protein molecules at high spatial resolution, which is highly desirable for analyzing proteins in living cells where a complex environment and molecule crowdedness often dictate the measurements. For AFM matrix mapping, we utilized an approach of combined AFM 2D matrix force pulling scanning and singlemolecule FRET imaging measurements. With the alignment of the AFM tip, the laser beam focus, and the target molecule in a co-axial configuration, the AFM tip and the single target molecule are both in the laser focus; however, the distance between the AFM tip and the target molecule can still be tens of nanometers. To ensure a single-molecule AFM-FRET measurement for the same target protein molecule, we use a new approach of AFM matrix pulling (or mapping) and simultaneous single-molecule FRET measurement. The typical size of the coated AFM tip apex is around 20–40 nm in diameter, and the targeted protein molecule is about 5–10 nm in diameter; therefore, a 20 20 nm2 area (about one pulling event area) is sufficient for each AFM-tip pulling to ensure a direct contact with the single molecule under the laser focal point. In a typical

1124 | Chem. Soc. Rev., 2014, 43, 1118--1143

Review Article

Fig. 6 (A) Single-molecule AFM-FRET ultra nanoscopy; the zoomed panel in the left represents the schematic diagram of a FRET dye-pair (donor– acceptor: Cy3–Cy5) labeled HPPK molecule tethered between the surface of a glass coverslip and a handle (biotin group plus streptavidin), and another biotin group is modified on the AFM tip. (B) Single-molecule fluorescence photon counting images of the donor (Cy3, left) and the accepter (Cy5, right). Each feature is from a single HPPK enzyme labeled with Cy3–Cy5 FRET dyes. (Adapted with permission from ref. 45. Copyright 2012 American Chemical Society.)

experimental protocol shown in Fig. 5 and 6, there is a 16 16 times pulling matrix within an area of 300 300 nm2, in which an AFM-tip force pulling event occurs in every 20 nm interval. Meanwhile, the single protein molecule can be reached under such a sampling matrix of every 20 20 nm2 within the laser focal point where one individual target molecule is located. Simultaneously, we record the single-molecule fluorescence intensities of the FRET pair of Cy3 and Cy5 by a two-channel photon-stamping module during the AFM matrix scanning. 2.5 Total internal reflection fluorescence microscopy imaging-guided confocal single-molecule fluorescence spectroscopy We have recently developed an integrated spectroscopy system combining total internal reflection fluorescence microscopy (TIRFM) imaging with confocal single-molecule fluorescence spectroscopy for two dimensional interfaces.155 This spectroscopy approach is capable of both simultaneous sampling of multiple molecules and in situ confocal fluorescence analyses on dynamics of specific individual molecules. Although both fluorescence confocal microscope imaging and total internal reflection fluorescence microscopy (TIRFM) imaging serve as essential approaches for observing single molecule dynamics with wide applications,3,4,9–13,15,17,19–39,44–47,49–53,55–61,85,86 they are complementary but also show their limitations when simultaneously requiring sampling view fields and sampling time resolutions. Simply, the TIRFM is for imaging a large number of molecules at a time but not for detailed spectroscopic analysis of each of the imaged molecules, whereas the confocal single-molecule fluorescence imaging is for the detailed analysis of the spectroscopic and dynamics of the targeted molecule but


View Article Online

Review Article

Chem Soc Rev


3. Single-molecule enzyme conformational dynamics in enzymatic reactions

Fig. 7 Calibration of TIRFM imaging-guided confocal single-molecule fluorescence spectroscopy. (A) Fluorescence image of the microspheres in CCD camera coordinates by TIRFM mode. (B) Fluorescence image of the microspheres in scanning translation stage coordinates by confocal mode. According to the special pattern, the microspheres a, b and c in (A) are the same as a, b and c in (B). To guide the confocal spectroscopic measurements of the spots of interest in the TIRFM frame, the coordinates of the spots of interest in the sample scanning stage frame are determined by matrix transformation.155 Feasibility of TIRFM imaging-guided confocal single-molecule fluorescence spectroscopy is demonstrated in panels C and D. (C) Image of Rhodamine 6G molecules in TIRFM mode. (D) The corresponding confocal single-molecule spectroscopy measurements on molecule a in (A). The fluorescence intensity decay (red) from single-molecule photon stamping recording is fitted with exponential decay (blue) with fit residual (green); the fluorescence lifetime of the single Rhodamine 6G molecule is 3.1 0.3 ns. Inset: the intensity trajectory of the single molecule (black).

not for sampling a large number of molecules at a time. Fluorescence confocal microscope is often equipped with avalanche photodiodes or single photon avalanche diodes, which are suitable for high time resolution measurements in time-resolved dynamics analyses (Fig. 7). However, the single pinpoint detection constrains the efficiency of detecting spatially and temporally randomly distributed single-molecule fluorogenic events, such as fluorogenic product turnovers of single-molecule enzymatic reactions. TIRFM cannot provide high time resolution due to two dimensional imaging. However, the much larger simultaneously sampling view fields dramatically improve the possibility of identifying the tethered enzymes only by the temporally random fluorogenic signal (Fig. 7). Our newly developed confocal-TIRFM imaging microscopy combines the advantages of wide field imaging of TIRFM with the advantage of high time resolution of confocal fluorescence spectroscopy, through (1) detecting reaction active sites and recording their coordinates by TIRFM, and (2) relying on the recorded coordinates to guide the confocal single-molecule spectroscopy measurements for individual pinpoints of interest (Fig. 7). This spectroscopy approach is capable of both simultaneous sampling of multiple molecules and in situ confocal fluorescence dynamics analyses of individual molecules of interest.


In the late 1990s, there was a significant application of singlemolecule spectroscopy for studying single-molecule enzymatic reaction dynamics,5,6 which opened up a new stage for singlemolecule enzymology. One of the significant works on singlemolecule enzymatic reaction dynamics studied cholesterol oxidase (COx) from Brevibacterium sp., a monomeric protein of 53 kDa, by probing the enzymatic reaction center fluorescence to read out the redox state changes under the redox enzymatic reactions. Flavin adenine dinucleotide (FAD) is the coenzyme and is non-covalently bound to the protein matrix at the active site. Each COx molecule contains only one FAD. In the oxidized form, FAD absorbs excitation photons at 450 nm and emits fluorescence at 520 nm; however, only the oxidized FAD is fluorescent. During cholesterol oxidation in an enzymatic turnover cycle, FAD is first reduced to FADH2, and then reoxidized back to its original form by oxygen. The reduced FADH2 is nonfluorescent. FAD toggles between fluorescent oxidized form and non-fluorescent reduced form in each enzymatic turnover cycle. Accordingly, the single-molecule fluorescence turns on and off as the redox state of the FAD toggles between the oxidized and reduced states. Each on–off cycle corresponds to an enzymatic turnover. The turnover trajectory contains detailed information about the chemical dynamics. Significant static disorder and dynamic disorder of the enzymatic reaction dynamics were observed; seemingly identical COx molecules having significantly different enzymatic reaction rates reflects static disorder in the FAD reduction process. For each individual enzyme, the enzymatic reaction rate also fluctuates significantly from time to time, which constitutes the dynamic disorder of the reaction dynamics. Furthermore, a memory effect in the enzymatic reaction dynamics was discovered and identified. It was found that a turnover time is not completely independent of the previous turnover times: a short turnover time is likely to be followed by a short turnover time, and a long turnover time is likely to be followed by a long turnover time, the so-called memory effect. Conformational changes have been considered to be the common origin for the other types of memory effects found in other monomeric enzyme proteins that are often called regulatory enzymes. In these enzymes, there is more than one conformational state available. The substrate can shift the enzyme from one state to another state by changing the free energy difference or the free energy barrier between the two states. These energy landscape changes can induce equilibrium shifting and rate changes of the regulatory transition.5,6 The static and dynamic disorders of the enzymatic reaction dynamics, especially the associated conformational dynamics, are extremely difficult, if not impossible, for the conventional ensemble-averaged measurements to analyze, and single-molecule spectroscopic analyses provide critical information for dissecting the new and important perspectives of the enzymatic reactions. In this session, we will focus our discussion on two enzymes that typically show significant conformational motions in each enzymatic reaction turnover cycles: T4 lysozyme with two domain

Chem. Soc. Rev., 2014, 43, 1118--1143 | 1125

View Article Online


Chem Soc Rev

Review Article

Fig. 8 Probing single-molecule T4 lysozyme conformational dynamics in the enzymatic hydrolysis of polymer sugar chain substrates. (A) Crystal structure of wild-type T4 lysozyme. The protein was labeled with Texas Red maleimide and tetramethylrhodamine iodoacetamide (or Alexa 488/Alexa 594 dye probes) by thiolation to Cys-54 and Cys-97. The significant advantage of our site-specific covalent dye labeling is that the attached donor–acceptor pair can sense the relative motion of the two domains in T4 lysozyme without perturbation of the enzymatic activity.9 (B) The T4 lysozyme was covalently linked to a hydrocarbonmodified glass coverslip by the bi-functional linker SIAXX (Molecular Probes, Inc.). At concentration of 109 M, the single-molecule enzyme molecules on surface at a density less than 1 molecule per mm2 so that the diffraction-limited spatially focused laser spot can conduct single-molecule FRET excitation and measurements of an individual molecule at a time. (C) Hydrolysis reaction of polysaccharide catalyzed by T4 lysozyme. (D) Representation of the enzymatic reaction active complex of polysaccharide substrate and T4 lysozyme. The negative surface charge is in red and positive surface charge in blue.

hinged bending motions and a kinase with flexible substrate loop binding motions. Fig. 8A shows the protein structure of the T4 lysozyme: the two domains are linked by a polypeptide helix as a hinge. The two domains presumably move around the hinge, and the domain motions result in open–close active site conformational states, for example, under substrate binding interactions in an enzymatic reaction. The relative open–close domain motions are the hinged bending motions. We will focus our discussions on (1) single-molecule FRET spectroscopy probing enzymatic conformational dynamics; (2) T4 lysozyme conformational changes under enzymatic reaction turnovers; (3) mechanistic understanding of the conformational dynamics in T4 lysozyme enzymatic reactions; (4) bunching effect in single-molecule T4 lysozyme nonequilibrium conformational dynamics under enzymatic reactions; (5) beyond the conventional correlation function analysis of complex conformational fluctuations: 2D regional correlation analysis of single-molecule time trajectories; (6) probing singlemolecule enzyme active-site conformational state intermittent coherence; (7) towards studying enzymatic reaction dynamics and conformational dynamics in living cells; and (8) singlemolecule manipulation and observation: AFM-FRET nanoscopy studies of enzymatic dynamics by actively manipulating protein conformational dynamics. 3.1 Single-molecule fluorescence resonance energy transfer spectroscopy probing enzymatic conformational dynamics There are various choices of the FRET donor–acceptor dye probes to be covalently attached to the Cys residues of the

1126 | Chem. Soc. Rev., 2014, 43, 1118--1143

proteins. By controlled reaction conditions and carefully chosen dye derivatives with appropriate functional linking groups, we were able to get purified single-dye or double-dye (donor–acceptor) labeled proteins in high yields. For example, we site-specifically labeled two cysteines with tetramethylrhodamine (TMR iodoacetamide) and Texas Red (Texas Red maleimide) on wild-type lysozyme protein through thiolation to probe the hinged bending motions of the T4 lysozyme intramolecular conformational changes in enzymatic reactions (Fig. 8). As a routine control experiment, the effects of fluorescence labels on enzyme proteins were typically evaluated by comparing enzymatic assays for native protein and labeled proteins. We have found that dye labeling can cause a wide range of perturbations to the enzyme activity: from no effect (for example, wild-type lysozyme) to up to more than 30% (for example, mutant lysozyme, E11A) predominantly depending on the sites of dye labeling. Our single-molecule spectroscopy studies on T4 lysozyme enzymatic conformational dynamics and catalytic dynamics have shed light on a number of fundamental perspectives about the enzymatic reaction mechanism: (1) the spatial and temporal ranges of the critical nanoscale conformational motions occurring during chemical processing by lysozyme; (2) the coordinate-specific characterization of these dynamic motions; (3) the characterization of the inter-domain and intra-domain conformational motions in the context of the single-molecule catalytic turnovers; (4) the detailed analysis of the catalytic complexes formed between the enzyme and the substrate on the dynamics of their association, dissociation and conversion to the product through enzymatic


View Article Online


Review Article

reaction turnovers; (5) the characterization of the order and disorder of the complex enzymatic active-site conformational dynamics; and (6) a quantitative understanding of the sensitivity and the distributions of the enzymatic rates to local environmental conditions. Using single-molecule spectroscopic approaches, we were able to characterize the spatial and temporal fluctuations of enzyme–substrate active complex formation, reaction, and product releasing; nanoscale conformational motions often have intrinsically disordered or flexible regions that exhibit large conformational changes during the reaction process which are critical to define the potential surface and activation barrier for the reaction, and eventually enable the catalytic reaction to occur.1–43 We are now able to analyze whether relatively slow (sub-millisecond to second) protein motions are the most critical in the reactions or whether fast motions are the key and identify the static and dynamic distributions of the protein motion rates. Furthermore, we are also able to identify which conformational change coordinate or coordinates are critical for the enzymatic reaction activity. The protein rotational motion and local electric field fluctuation dynamics can be studied by fluorescence anisotropy from intrinsic fluorophores or a dye with high sensitivity to the local electric field. The enzymatic conformational changes are often associated with complex dynamics that are typically difficult if not impossible for ensemble-averaged experiments to identify and analyze. For example, the static disorder and dynamic disorder in the inhomogeneous conformational fluctuation rates or the catalytic turnover rates are hard to be identified and characterized.1–43 Single enzyme–substrate complexes can be visualized through monitoring single-molecule time trajectories of binding motions in different local environments in combination with statistical model analyses5,6,11,12,29,36 and computational simulations.9 Molecular-level understanding of the effects of the local nanoscale environment is crucial for understanding and modeling the complex enzymatic reactions in cell wall hydrolysis that are present in a wide range of temporal and spatial heterogeneities. To dissect these complex dynamic behaviors, single-molecule spectroscopy has been proven to be highly informative.9,24,29,44–47,49–53,55–61,85,86

3.2 T4 lysozyme conformational changes under enzymatic reaction turnovers The domain motion of T4 lysozyme is rather complex and contains motions besides hinge-bending. It is reasonable to assume that the hinge-bending motion in nature involves multiple coupled nuclear coordinates that can be projected to a nuclear coordinate associated with the a-helix. Based on our study9 and a previous156 molecular dynamics simulation of wild-type T4 lysozyme in solution, the distance change between two dye-tethered cysteine residues is from 30.5 Å to 35 Å, i.e., the donor–acceptor distance change is about 4.5 Å. We esti¨rster distance R0 (ref. 24) of a TMR/Texas Red pair mated the Fo to be about 50 5 Å. The lengths of the two covalent linking groups of the donor and acceptor dipoles, which have an actual


Chem Soc Rev

donor–acceptor distance change of 5.5 Å,9 can cause a change of 2–3 times in donor fluorescence intensity.9,29,44,50,59 To demonstrate the feasibility of applying FRET to probe the conformational changes of T4 lysozyme proteins under binding and unbinding hinge-bending motions, we measured the ensemble-averaged FRET emission spectra from a donor– acceptor-labeled T4 lysozyme mutant (E11A), both with and without the substrates (Fig. 9). The T4 lysozyme mutant is ideal for the ensemble-averaged control experiment because it is only involved in binding and unbinding interactions with the substrate but has no catalytic reactivity. The ratio of spectral intensities of the donor emission vs. acceptor emission significantly increases on enzyme’s attaching to the substrate. The ensemble fluorescence intensity measurement indicates that there are significant conformation changes of T4 lysozyme upon binding to the substrate although it is essentially a

Fig. 9 (A) A modified Michaelis–Menten mechanism specifying that the enzyme (E) forms a non-specific binding complex (ES) with the substrate (S), and then the ES transforms to the active complex (ES*) mostly through conformational changes, and then a chemical reaction (EP) followed by product release (E + P). Note that the enzymatic reaction equation in this figure only presents a conceptual interpretation of the catalytic mechanism. Our later discussion will specify a detailed characterization of the mechanism. (B) Ensemble-averaged control FRET experiment of probing T4 lysozyme conformational open–close motions, opening up to intake the substrate and binding down to form the active complex (ES*), in the hydrolysis of an E. coli B cell wall. There are no significant conformational changes in reactions and product release; therefore, for each enzymatic reaction cycle, an open–close enzymatic conformational motion is involved. The fluorescence spectra of Alexa 488/Alexa 594 labeled T4 lysozyme mutant (E11A) excited at 488 nm are shown. The blue and red lines are the fluorescence spectra of the enzyme in solution without and with substrate E. coli B cell walls present, respectively. The two spectra are normalized to the same maximum point, to aid the view. It is clearly shown that the FRET efficiency significantly decreased when the enzyme is under enzymatic reaction with the substrates. The decrease of the FRET efficiency is because the two domains of the enzyme are in open–close conformational motions spending more time in average in the conformations that the two domains are farther away in the open state. (C) Based on our single-molecule FRET and computational simulation, the active site of T4 lysozyme opens up for the substrate to diffuse in (E + S - ES) and closes down to form the active substrate–enzyme complex (ES - ES*) ready to react. The specific conformational changes in each enzymatic reaction cycle, including productive and non-productive cycles, are the origin of the measured FRET changes.

Chem. Soc. Rev., 2014, 43, 1118--1143 | 1127

View Article Online


Chem Soc Rev

Fig. 10 Single-molecule recording of T4 lysozyme conformational motions and enzymatic reaction turnovers of hydrolysis of an E. coli B cell wall in real time. (A) Fluorescence image (20 mm 20 mm) of single T4 lysozyme molecules tethered to the hydrocarbon-modified glass surface of a coverslip under pH 7.2 aqueous buffer solution. The fluorescence emission was split by a dichroic beam splitter (595 nm long-pass). The donor (left panel) and acceptor (right panel) emissions were detected separately by a pair of avalanche photodiode detectors after passing through a 570 nm band-pass filter (20 nm bandwidth) and a 615 nm long-pass filter, respectively. The two images were taken from the same area of a sample with an inverted fluorescence microscope by rasterscanning the sample with a focused laser beam of 100 nW at 532 nm. Each individual peak is attributed to a single T4 lysozyme molecule. The intensity variation among the molecules is predominantly due to smFRET. (B) A pair of trajectories from a fluorescence donor tetramethylrhodamine (blue) and acceptor Texas Red (red) pair in a single T4 lysozyme in the presence of E. coli cells of 2.5 mg mL1 in pH 7.2 buffer. Anticorrelated fluctuation features are evident. (C) The correlation functions (C(t)) of donor (hDId(0)DId(t)i, blue), acceptor (hDIa(0)DIa(t)i, red), and donor–acceptor cross-correlation function (hDId(0)DIa(t)i, black), deduced from the single-molecule trajectories in (A). They are fitted with the same decay rate constant of 180 40 s1. A long decay component of 10 2 s1 is also evident in each autocorrelation function. The first data point (not shown) of each correlation function contains the contribution from the measurement noise and fluctuations faster than the time resolution. The correlation functions are normalized, and the hDIa(0)DIa(t)i is presented with a shift on the y axis to enhance the view.9 (D and E) A single-molecule control experiment results without substrate and enzymatic reaction turnovers. It is evident that although the donor and acceptor signal trajectories are measurable and similar to the trajectories measured under enzymatic reaction conditions, the correlation functions show distinctive features of showing no protein fluctuation dynamics resolved.

qualitative confirmation of FRET existence under the interaction between the enzyme and the substrate. The fluorescence intensity trajectories of the donor (Id(t)) and the acceptor (Ia(t)) give autocorrelation times (Fig. 10) indistinguishable from fitting an exponential decay to the

1128 | Chem. Soc. Rev., 2014, 43, 1118--1143

Review Article

autocorrelation functions, hDId(0)DId(t)i and hDIa(0)DIa(t)i, where DId(t) is Id(t) hIdi, hIdi is the mean intensity of the overall trajectory of a donor, and DIa(t) has the same definition for an intensity trajectory of an acceptor. In contrast, the crosscorrelation function between the donor and the acceptor trajectories, hDId(0)DIa(t)i, is anticorrelated with the same decay time (Fig. 10C), which supports our assignment of anticorrelated fluctuations of the fluorescence intensities of the donor and the acceptor to the smFRET process. Our further control experiment of nanosecond anisotropy proved that the tethered enzyme can freely rotate and has a minimum perturbation from the hydrocarbon-modified glass surface (Fig. 8B).12 We used femtosecond laser pulse excitation in the single-molecule anisotropy measurements to probe the nanosecond molecular rotational motions when the single T4 lysozyme enzyme molecules were confined in the agarose gel, we identified that the enzyme molecules are confined but free in rotational motions. We observed autocorrelation rate constants that differed by a factor of 2 over the pH range from 7.2 to 6.0: at pH 7.2, the average decay rate constant was 160 15 s1, and at pH 6.0, the rate constant was 80 10 s1. This 2-fold decrease in the decay rate constant is consistent with the enzymatic activity decrease measured by ensemble-averaged assays at pH 7.2 and 6.0.9,29 It has been known that lowering the pH alters the surface charge density of the protein by protonation of the surface carboxylic groups and histidine residues (when the substrate is present) and, therefore, perturbs the protein conformation and ultimately the overall enzymatic hydrolysis reaction rate. Single-molecule smFRET fluorescence trajectories contain detailed information about the conformational fluctuation dynamics associated with the enzymatic turnovers. The upper panel in Fig. 11 shows an expanded portion of a trajectory (middle panel) recorded from the donor fluorescence of a single-pair donor–acceptor labeled enzyme with the substrate present, and the significant signal intensity amplitude wiggling changes are measurable beyond the experimental signal-tonoise ratio. We have systematically carried out a singlemolecule spectroscopy control experiment under the assay conditions: single-molecule imaging configuration and parameters remaining the same but only the enzyme is donoronly-labeled instead of FRET donor–acceptor labeled. In this control experiments, single-molecule images from detecting the donor emission signals can be obtained but no sensitivity of probing the conformational changes as there is no FRET pair to probe the conformational hinged bending motions under enzymatic reactions. The lower panel shows a portion of a donor-fluorescence trajectory recorded from a donor-only labeled T4 lysozyme under the same enzymatic reaction assay conditions. The large-amplitude, lower-frequency wiggling of the donor fluorescence intensity in the upper panel is largely absent from the trajectory in the lower panel showing a stable averaged intensity with photon counting noises from the single donor molecule alone. The inset in Fig. 11A (middle panel) shows a bimodal fluorescence intensity distribution that reflects the open and closed conformational states of a T4


View Article Online


Review Article

Chem Soc Rev

with the systematic control results of the ensemble-averaged FRET measurements and the single-molecule FRET intensity fluctuation dynamics analysis made with and without a substrate.9 We have carried out systematic control experiments at both ensembleaveraged and single-molecule levels. Further evidence that we are measuring the hinge-bending motion comes from evaluating autocorrelation functions calculated from the single-molecule donor and acceptor trajectories when changing the laser excitation intensity, the pH, and the substrate concentration.9 We did not observe a dependence of fluctuation correlation time on the excitation rate, indicating that the fluctuations were spontaneous rather than laser-driven. However, we did observe autocorrelation rate constants that differed by a factor of 2 over the pH range from 7.2 to 6.0 as we have discussed above.9 3.3 Mechanistic understanding of the enzymatic active-site conformational dynamics in T4 lysozyme catalytic reactions

Fig. 11 (A) Real-time observation of single-lysozyme conformational motions and enzymatic reaction turnovers during the hydrolysis of bacterial cell walls. The data in the three panels were recorded at 0.65 ms per channel for the same reaction conditions. The upper panel shows an extended portion of the middle-panel trajectory of the donor fluorescence of a donor–acceptor labeled single lysozyme. Intensity wiggles in the trajectory are evident. The lower panel shows a portion of a trajectory recorded from a donor-only-labeled protein. The fluorescence intensity distributions derived from the two trajectories are shown in the insets of the middle and lower panels. The solid lines are fits using single and bimodal Gaussian functions, respectively. The formation of ES and ES* is associated with significant domain breathing motions along the a-helix ‘‘hinge’’ causing the large amplitude wiggles and the bimodal amplitude distribution. From this it can be seen that smFRET measurements are effective real-time probes of the enzymatic reaction dynamics.9 (B) Activecomplex, ES*, formation time (ton) distribution deduced from a single T4 lysozyme fluorescence trajectory under enzymatic reactions. The ton is the duration time of each wiggling of the intensity trajectory above a threshold. The threshold is determined by the 50% of the bimodal intensity distribution as mentioned in (A). The mean open time, htoni, is 19.5 2 ms and the standard deviation of the open time is 8.3 2 ms. The measured ton reflects the time for the formation of the active enzyme–substrate complex (ES*). The enzyme active site opens up to take substrate in and form the non-specific binding complex (ES) and closes down to form the active complex (ES*). The single-molecule open–close hinge-bending motions are measured by single-molecule FRET spectroscopy and recorded in the FRET time trajectories.

The single-molecule experimental trajectory data (Fig. 11A) allow us to have a molecular-level understanding of the T4 lysozyme enzymatic reaction mechanism. The donor–acceptor distance and the donor fluorescence intensity increase when the active site opens up to form a nonspecific binding complex (ES) with the substrate, corresponding to the process of E + S ES. The donor–acceptor distance and the donor fluorescence intensity decrease when the active site closes to form an active complex (ES*), corresponding to the process of ES - ES* (Fig. 8–12). There are no measurable smFRET changes and significant conformational open–close motions in the process reaction to convert the substrate (S) to the product (P), ES* - EP or in the product-releasing process of EP - E + P (Fig. 9 and 12). The formation of ES and ES* is intrinsically associated with significant domain breathing-type hinge-bending motions along the a-helix connecting the two enzyme domains (Fig. 9 and 12), and the active-site domain open–close motions are probed in real time by recording single-molecule smFRET trajectories that record the formation times, topen, of enzymatic intermediate ES and ES* states from the single-molecule enzymatic turnover trajectories (Fig. 11A).9 Fig. 11B shows a Gaussian-shaped distribution of the open-time (topen) deduced from a single-molecule trajectory. The first moment of the distribution, htopeni = 19.5 2 ms, corresponds to the mean time of the processes of E + S - ES - ES*, as shown in Fig. 11. The qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi standard deviation of the distribution, Dtopen2 ¼ 8:3 2 ms,

lysozyme. By contrast, only a Gaussian-like-shaped distribution was deduced from the trajectory of a donor-alone labeled single enzyme molecule. This indicates that only fast fluctuations and uncorrelated noise were recorded, since the donor dye alone is not sensitive to the enzyme’s open–close hinge-bending motions (Fig. 11A, lower panel). We have attributed each wiggle to a hinge-bending motion in either an enzymatic productive reaction or nonproductive binding and release of the substrate.9 The donor fluorescence intensity increases as the active site opens due to substrate insertion and decreases as the active site closes to form an active enzyme–substrate complex. This attribute is consistent

reflects the distribution bandwidth. For the individual T4 lysozyme molecules examined under the same enzymatic reaction conditions, we found that the mean and the standard deviation of the single-molecule topen distributions are rather homogeneous within the error bars. The hinge-bending motion allows sufficient structural flexibility for the enzyme to optimize its domain conformation: the donor fluorescence essentially reaches the same intensity in each turnover, reflecting the domain conformation reoccurrence. The non-equilibrium conformational motions in forming the active enzymatic reaction intermediate states intrinsically define a recurrence of the essentially similar potential surface for the enzymatic reaction to occur, which


Chem. Soc. Rev., 2014, 43, 1118--1143 | 1129

View Article Online


Chem Soc Rev

Fig. 12 (A) An illustration of the ton measured from the donor intensity changes. (B) Molecular dynamics simulation of T4 lysozyme hinge-bending motion. Three conformation trajectories of T4 lysozyme in solution at room temperature are presented: free enzyme without substrate (E), active enzyme-polysaccharide complex (ES*), and nonspecific binding of polysaccharide (E + S - ES). (left panel) The time trajectory of the distance between two –SH of Cys-54 and Cys-97 of free enzyme. (middle panel) The time trajectory of non-active enzyme–substrate formation dynamics as the polysaccharide moves into the active site of the T4 lysozyme. (right panel) The time trajectory of the distance between two –SH of Cys-54 and Cys-97 of the T4 lysozyme–polysaccharide active complex.9 Enzyme structure (E) was taken from the 1.7 Å X-ray crystallographic structure (PDB entry 3LZM) and included 152 water molecules. The system was placed in a periodic cube of 73.34 Å per side and filled with 11 948 SPCE water molecules, including a section of the substrate (ES*). A six-unit oligosaccharide consisting of alternating N-acetylmuramic acid (NAM) and N-acetylglucosamine (NAG) was positioned in the active site with the aid of superimposing the lysozyme mutant adducted with the substrate cleaved from the cell wall of E. coli (PDB entry 148L). (C) Two conformation states of T4 lysozyme protein surface presentations based on the crystal structures of wt and a mutant T4 lysozyme, showing both closed and open conformations.

represents a time bunching effect in the enzymatic reaction conformational dynamics.9,29,49,54,55,57 It is likely that the substrate binding selectively shifted the equilibrium among the fluctuating conformations of the enzyme which energetically and dynamically regulated the hinge-bending motions into a specific conformational fluctuation time range. The results of the MD simulation9 suggest that the dominant driving force for E + S - ES is the positive surface charge of the enzyme from surface amino acid residues (arginine and lysine) interacting with the negatively charged polysaccharide substrate. The driving force for ES - ES* includes the formation of six hydrogen bonds in the active site of ES*. 3.4 Correlated model analysis of the conformational change energy landscape based on the single-molecule conformational dynamics and MD simulation Our combined analyses based on single-molecule FRET spectroscopy and computer simulation provide an opportunity to actually

1130 | Chem. Soc. Rev., 2014, 43, 1118--1143

Review Article

characterize the energy landscape of the T4 lysozyme enzymatic active-site conformational fluctuations and the catalytic reaction dynamics. We model the hinge-bending motion associated with interactions between the enzyme and the substrate as a classical particle one-dimensional multiple-step random walk in the presence of a force field.100 Here, the n-step random walk is represented as {n(t)}, where n(t) is the step index. The kf and kb are the forward and backward step rate constants, respectively. The driving force would tend to make kf > kb. The position distribution density function Pn(t) can be calculated24 by dPn ðtÞ ¼ kf Pn1 ðtÞ þ kb Pnþ1 ðtÞ ðkf þ kb ÞPn ðtÞ. dt Assuming the random-walk step-size of L and a drifting distance of Xn = nL, we have hDXn(t)2i = L2(kf + kb)t and hXn(t)i = L(kf kb)t. Considering the one-dimensional random walk and approximate Gaussian-shape of Pn(t),9 we have D = hDXn(t)2i/2t = L2(kf + kb)/2 and hvi = hXn(t)i/t = L(kf kb), where hvi is the mean drifting velocity of the conformational change along the a-helix ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi rD E DXN ðtÞ2 coordinate. With the approximation of ¼ hXN ðtÞi ffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Dtopen2 in the long-time limit, where N is the index of the topen qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 ffi Dtopen2 hXN ðtÞi final state, we have D ¼ , where D is the 3 2 topen diffusion coefficient. The total drifting distance of the conformational open–close motion, hXN(t)i, is about 9 Å, based on our MD simulation (Fig. 12). The mean open-time, htopeni, and the qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi standard deviation of the open-time distribution, Dtopen2 , are measured to be 19.5 2 ms and 8.3 2 ms, respectively. Therefore, the mean drifting velocity, hvi, of the conformational change in E + S - ES - ES* is hvi = hXN (t)i/htopeni = 4.6 106 cm s1, and the diffusion coefficient is, therefore, qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 ffi Dtopen2 hXN ðtÞi ¼ 3:8 1014 cm2 s1 . D¼ 3 2 topen We further characterized the energy landscape of the hingebending conformational change dynamics by estimating the minimum number of the intermediate conformational states in the complex formation and by calculating the averaged rates of forming these conformational states. We have D/hvi = L[(kf + kb)/ 2(kf kb)], and L = 2D/hvi = 1.6 Å, when kb - 0; therefore, the minimum number of conformations is m = hXN (t)i/L = 5.6. This result (m > 2 at the limit of kb - 0) suggests that there are more than two conformational intermediate states in addition to ES and ES*. With the assumption of kb - 0, the friction coefficient can be estimated using the Einstein relationship x = kT/D giving x = 1.1 erg s cm2. The energy consumed by friction in the drifting process is Ef = x hvihXN (t)i = kThvihXN (t)i/D = 4.5 1020 J. From MD simulation of the E state and the ES* state energies, our best estimate of the total energy change between the two states is 18 kcal mol1 (1.25 1019 J per molecule) which comprises the energy gains from electrostatic and van der Waals interactions and hydrogen-bond formation. Therefore, we


View Article Online


Review Article

estimated that 36% of the total energy change between E and ES* is spent on the friction along the reaction coordinate. Since the total energy change between the E state and the ES* state is about 18 kcal mol1, assuming six intermediate states would result in an average energy difference of 3 kcal mol1 for each associated conformational state along the a-helix coordinate during the hinge-bending motion. We postulate that the activation energy associated with the kf of the forward step is about 0–5 kcal mol1, considering the slow forward rate and the entropy decrease in the complex formation process.9,29,55 It is reasonable to assume the energy potential surface along the conformational change nuclear coordinate to be parabolic for each intermediate state (Fig. 13). Therefore, the averaged force constant of the potential surface for each intermediate state is calculated to be 1.710 kcal mol1 Å2.9,29,55 The unique and detailed information about the enzyme conformational dynamics has helped us to set up a new stage for combining analyses of three powerful and highly complementary approaches: the experimental single-molecule time trajectories and

Fig. 13 Based on the results of the single-molecule spectroscopy measurements, an attempt was made to estimate the energy potential surface of the T4 lysozyme–substrate complex formation process. The conformational change dynamics of multiple intermediate states is analyzed based on a one-dimensional random walk model coupled with the parameters yielded from our single-molecule experimental spectroscopy and MD simulation. The conformational motion in each enzymatic turnover cycle comprises approximately six intermediate states based on the model analysis. Since the total energy change between E and ES* states is about 18 kcal mol1, assuming six intermediate states would result in an average energy difference of 3 kcal mol1 for each associated conformational state along the a-helix coordinate during the hinge-bending motion. We postulate that the activation energy associated with the kf of the forward step is about 0–5 kcal mol1, considering the slow forward rate and the entropy decrease in the complex formation process. It is reasonable to assume the energy potential surface along the conformational change nuclear coordinate to be parabolic for each intermediate state. Therefore, the averaged force constant of the potential surface for each intermediate state is calculated to be 3.4–20 kcal mol1 Å2.9,29,55 (Adapted with permission from ref. 55. Copyright 2011 American Chemical Society.)


Chem Soc Rev

dynamics, the computational molecular dynamics (MD) trajectories and dynamics, and various condensed-phase theoretical model analyses. Using the combined analyses, a number of critical physical properties of the enzymatic reaction dynamics and associated conformational dynamics can be defined and analyzed quantitatively, including the microscopic conformational change mean drifting velocity, diffusion coefficient, friction coefficient, and Ef. Furthermore, the new combined approach also provides information about the existence of the multiple intermediate conformational states involved in the enzymatic active complex formation and a detailed characterization of the energy landscape of the complex formation process. 3.5 Bunching effect in single-molecule T4 lysozyme non-equilibrium conformational dynamics under enzymatic reactions Experimentally, the Gaussian-like distribution of the topen and the ramping changes of intensity at a millisecond time-scale in the single-molecule fluorescence time trajectories (Fig. 11) suggest that the protein hinge-bending conformational changes are associated with multiple intermediate conformational states and show the time bunching effect, which is consistent with the model analysis result of involvement of about 5.6 intermediate states in forming the enzyme–substrate complex (ES*).2–6,9,29,55,57 Here, we use the term ‘‘bunched’’ to specify the fact that the distribution is Gaussian-like with finite first and second moments, which is in contrast to the exponential distribution of a Poisson rate process. The Gaussianlike distribution of the topen cannot be characterized as an overall Poisson stochastic rate process but rather as the result of convolution of consecutive multiple Poisson rate processes. From the enzyme–substrate interaction perspective, we note that the bunching effect is intrinsic because the dominant nuclear coordinate is essentially the same for the bending motion that opens and closes the active site and, most likely, is associated with the a-helix hinge of the T4 lysozyme and relatively identical substrate–enzyme electrostatic force of the induced conformational changes. The bunching effect, a dynamic behavior observed for the catalytic hinge-bending conformational motions of T4 lysozyme, is attributed to a convoluted outcome of multiple consecutive Poisson rate processes that are defined by protein functional motions under substrate–enzyme interactions; i.e., convoluted multiple Poisson rate processes give rise to the bunching effect in the enzymatic reaction dynamics.54,55,57 It has been extensively reported that the enzyme may experience a set of intermediate states or transient states before reaching a reactive enzyme–substrate complex state.2–6,9,20–22,26,29–31,35–37,39,44,49–60 In complex rate processes, such as a non-equilibrium protein reaction process, with the local environmental fluctuation, nonexponential dynamics can be observed in reaction dynamics and conformational dynamics.2–6,9,20–22,26,29–31,35–37,39,44,49–60 To further specify our understanding of the bunching effect in enzymatic active-site conformational dynamics, we have simulated the T4 lysozyme enzyme–substrate active-state formation time, topen, probability distributions measured in our

Chem. Soc. Rev., 2014, 43, 1118--1143 | 1131

View Article Online


Chem Soc Rev

Review Article

single-molecule spectroscopic experiments for the T4 lysozyme enzymatic reaction. We proposed a modified Michaelis–Menten mechanism of the enzymatic reaction (Fig. 9, 11 and 12) based on the experimental data and a random walk model analysis discussed in the above section.9,29,55,57,100 The probability function P(Tn) (for six intermediate steps, n = 1, 2, 3, 4, 5, 6) of the formation times was then obtained from the convolution model and a computational simulation. ESn represents a non-specifically bound complex state and ES* represents a specifically bound complex ready to react. The enzymatic reaction is primarily driven by the electrostatic attraction between the positively charged surface of T4 lysozyme’s surface amino acid residues (arginine and lysine) and the negatively charged polysaccharide substrate. Being propelled by the electrostatic attraction, the positively charged front surface of the T4 lysozyme and the negatively charged substrate approach each other to form an enzyme–substrate non-specifically bounded complex state, ESn. During the enzyme–substrate interaction, six hydrogen bonds9,29,55,57,96,99 form at the enzyme active site with the substrate to form the active complex, ES*. Considering the perturbation of the local environment and thermal effect, the step between any adjacent states (for example, from state ESn1 to state ESn or from the last intermediate state to the active state ES*) is a stochastic process, and the step time obeys Poisson statistics. For the convoluted intermediate state transition, one can get the probability function P(Tn) from the integral algorithm of eqn (3). To calculate the convolution of functions f (t) and g(t), the particular integral transform is ðt ð f gÞðtÞ ¼ f ðnÞ gðt nÞ dn (3) 0

The general probability function is deduced to be PðTnÞ ¼ An

tn1 ½expðt=tÞ ðn 1Þ!

(4)

where n (1, 2, 3,. . .,N) is the index of the intermediate steps; t is the mean step time of an intermediate state through a singlestep rate process. The probability functions of the convoluted multiple intermediate states (or transient states) presented here can also be applied to other single-molecule enzyme conformational dynamics. Based on the function P(Tn) (n = 1, 2, 3,. . .,6) of eqn (4), the distribution of conformational motion time for multiple consecutive intermediate steps in forming the active complex of ES* is simulated. The histogram of the simulated formation times for only one intermediate step shows a typical Poisson distribution (Fig. 14) and a characteristic wing feature in the 2D joint probability distribution (Fig. 14), implying that there is no correlation and bunching effect in the formation times. The simulated conformational motion times involved in multiple intermediate states unambiguously show bunching effect as evident by the 2D joint probability distributions (Fig. 14). The observed bunching effect implies that the conformational motion time tends to be distributed in a finite and narrow time window as a Gaussian-like distribution with defined and comparable first and second moments. Interestingly, the

1132 | Chem. Soc. Rev., 2014, 43, 1118--1143

Fig. 14 Simulated distribution of formation times and corresponding two dimensional joint probability distributions of adjacent formation times for different intermediate steps. As a single step, stochastic and unbunched formation times show exponential distribution and only show a wing structure since that one step is a Poisson process and there is no bunching effect. However, for the multiple steps such as 2, 4, 6 steps, non-exponential distributions of the probabilities and the bunching features are increasingly clear, implying the bunching nature in the open–close conformational motion times. (Adapted with permission from ref. 41. Copyright 2010 American Chemical Society.)

experimental results show that the values of the first and second moments are comparable, which show the characteristics of multiple-step protein domain motion dynamics in T4 lysozyme enzymatic reactions, and the time scale is consistent with the overall enzymatic reaction rates measured in conventional ensemble-averaged experiments.157 The existence of the bunching effect in the conformational motion time has significant implications: the hinge-bending open–close conformational motions optimize the physical and chemical flexibilities of the enzyme to similar domain configurations for forming the same ES* complex in the enzymatic reaction turnover cycles. The physical picture strongly suggests that the conformational dynamics of T4 lysozyme shows a characteristic behavior of enzyme conformation selection dynamics driven by substrate–enzyme interactions. In a Poisson rate process, there should be no bunching among the stochastic conformational motion times. However, a nonequilibrium rate process through a sequence of consecutive Poisson processes with comparable rates eventually produces a bunching effect within the overall time lapse for the overall multiple-step rate process (Fig. 14). Typically, the physical nature of the bunching effect is associated with the functional conformational motion mechanism2,4,9,22,29,36,55,57 associated with non-equilibrium conformational fluctuations.9,22,24,29,36,55,108,109,130,131,133,136,137,139–142,146,158–161 The characteristics of the non-equilibrium conformational fluctuation dynamics are, for example here for T4 lysozyme, experimentally observed by the conformational open–close motions and the Gaussian-like substrate–enzyme formation time distributions.9 The conformational dynamics is significantly regulated by the interactions between the T4 lysozyme and the substrate in terms of electrostatic attraction and hydrogen bonding interactions. Our simulated data for six intermediate


View Article Online


Review Article

steps show essentially the same ES* formation time distribution with a Gaussian-like profile and the mean value of ton compared with the experimental ton distribution (Fig. 11B). Typically, the functional conformation selection mechanism applies when bunching effects in the conformational dynamics exist. As strong experimental evidence of ordered molecular conformational motions emerges from conformational fluctuations under enzymatic reaction conditions, bunching effect from enzyme conformational motions has significant and general biological relevance and consequences associated with some profoundly important biological properties, including biological oscillations and self-organizations, temporal and spatial functionality and complexity, and biological rhythms. It is the substrate–enzyme binding and the substrate– enzyme complex formation that serve as a driving force and negative entropy source for the coherent dynamics at the singlemolecule level. Accordingly, the conformational recurrence is driven by recurrence of the local environment, such as electrostatic field, electrostatic interactions, and hydrophobicity at the active site of the enzyme.24,162–164 Overall, in the absence of substrates, the enzyme explores a wide range of conformational space and consequently undergoes accessible conformational changes at a broad time scale, and when substrates are present, the conformational changes are regulated by the interactions between the enzyme and the substrate, so the time scale of these conformational changes is bunched up to a narrow scale. The conformational selection mechanism dominates the protein conformational fluctuation at non-specific binding between the substrate and the enzyme. Ligand binding to the active site changes the enzyme conformational state distributions to a much selected narrow subset, whereas the induced fit mechanism dominates the specific binding between the substrate and the active site of the enzyme in forming the active substrate–enzyme complex ready to react, which results in the appearance of a temporally bunched behavior in conformational dynamics. Our attribution is also consistent with recent reported works on the effects of enzyme–substrate interactions on conformational transitions of other enzymes.141,165–167 Through hydrogen bonding, electrostatic interactions, and solvent fluctuations, the substrates play an induced-fit role in regulating conformational change fluctuation patterns and rates in a thermal fluctuation local environment at the enzyme active site. Furthermore, it seems that non-functional conformational changes are similar, but functional conformational changes are selected and regulated by the enzymatic reaction conditions. 3.6 Beyond the conventional correlation function analysis of complex conformational fluctuations: 2D regional correlation analysis of single-molecule time trajectories There is a significant advantage of measuring the intra-molecular conformational changes of proteins by using single-molecule FRET spectroscopy: the corresponding changes of the FRET efficiency are typically at a level below 30%, a significant contrast to the often 100% changes observed for the intra-molecular DNA and RNA conformational changes as well as some of the inter-molecular conformational changes. The relatively small FRET efficiency


Chem Soc Rev

changes are due to the intrinsically small scale of intramolecular conformational changes. Consequently, a typical anti-correlated FRET D–A intensity fluctuation may be intermittently buried under the correlated or non-correlated fluctuations originated from the local environment fluctuations. For example, measurement noise, fluctuations beyond the measurement time resolution, and the thermal fluctuations of the local environment often intermittently dominate segments of a trajectory to show correlated or non-correlated fluctuations.24,44,55,57,61,168 Practically, an overall calculation of the time-correlation function may not reveal the intermittently appearing anti-correlated FRET fluctuation, or even show correlated thermal fluctuations.4,8–13,15,17,19–39,44–47,55,57 We have developed a two-dimensional correlation analysis approach to reveal the anti-correlated FRET donor–acceptor (D–A) fluorescence fluctuations from a correlated or non-correlated noise and thermal fluctuation background.24,44,55,57,61,168 Our 2D regional correlation analysis method is capable of identifying the FRET D–A anti-correlated fluorescence intensity fluctuations at any segments of an experimental fluctuation time trajectory, which is not possible for a conventional correlation analysis to average over across the whole time trajectory. Using this new approach, we were able to map out any defined segments along a fluctuation trajectory and determine whether they are correlated, anti-correlated, or noncorrelated, after which, a detailed cross correlation analysis can be applied for each specific segment for a fluctuation dynamics analysis.5,6,11,12,24,36,44,55,57 The cross-correlation evaluates the time-dependent strength between two fluctuating variables.11,12,24,36,44,55,57 The crosscorrelation (Ccross(t)) functions are defined by eqn (5) and (6). Ccross(t) = hDA(0)DB(t)i/hDA(0)DB(0)i = h(A(0) hAi)(B(t) hBi)i/h(A(0) hAi)(B(0) hBi)i (5) When A = B, we have the autocorrelation function Cauto(t) = hDA(0)DA(t)i/hDA(0)2i = h(A(0) hAi)(A(t) hAi)i/h(A(0) hAi)2i

(6)

where A(t) and B(t) are the signal variables measured in time trajectories {A(t)} and {B(t)}. hAi and hBi are the means of the fluctuation trajectories of {A(t)} and {B(t)}, respectively. In spectroscopic fluctuation analyses, {A(t)} and {B(t)} can be the time trajectories of spectral intensity, spectral mean, fluorescence polarization, photon counts, and other physical parameters. Mathematically, a time-correlation function for continuous or discrete fluctuation trajectory {A(t)} is calculated by Cauto ðtÞ ¼ hDAð0ÞDAðtÞi=hDAð0Þ2 i ð ð ¼ dtðAðtÞ hAiÞðAðt tÞ hAiÞ dtðAðtÞ hAiÞ2 ¼

X

.X ðAðtÞ hAiÞðAðt tÞ hAiÞ ðAðtÞ hAiÞ2 (7)

The cross-correlation analyses are methods to study the time-dependent behaviors of fluctuating signals.

Chem. Soc. Rev., 2014, 43, 1118--1143 | 1133

View Article Online


Chem Soc Rev

Review Article

The primary analytical approach of the 2D regional correlation analysis is to calculate a two-dimensional cross-correlation function amplitude distribution (TCAD). In this analysis, a start time and a stop time, tstart and tstop, were scanned for calculation of the cross-correlation function from a two-band signal intensity time trajectory, {I1(t)}, {I2(t)}. The two scanning parameters, tstart and tstop, define the start time (tstart) and the time lapse (from tstart to tstop) of a cross-correlation function calculation window along a two-band fluctuation signal trajectory. This 2D calculation gives a crosscorrelation for defined segments from tstart to tstop as ð tstop I1 ðtÞI2 ðt tÞdt Ccross ðt; tstart : tstop Þ ¼ ¼

tstop X

tstart

I1 ðtÞI2 ðt tÞ

(8)

tstart

The window of tstart to tstop is scanned in a range through the intensity trajectories. A cross-correlation function is calculated from a two-band fluctuation trajectory for each scanned pair of tstart to tstop. The initial amplitude of C(t, tstart:tstop) was represented by the difference between the first n points and the next n + m points from t = 0: z = {hC(1:n)i} {hC(n + 1:n + m)i}

(9)

The indices n and m define the precision of the calculated initial amplitude, z, of the correlation function. In our analysis, we chose n = m = 3, which is sufficient to identify a reliable value of z from the calculated cross-correlation function. As a function of tstart and tstop, the value of z is plotted as a two-dimensional map of tstart to tstop. A hot color represents the positive amplitude of C(t) and a cold color represents the negative amplitude of C(t). A positive amplitude indicates correlation, and a negative amplitude indicates anti-correlation. The significant advantage of TCAD is that both the correlated and anti-correlated spectral intensity fluctuations can be identified pixel-by-pixel for each pair of tstart to tstop based on the calculated cross correlation function pixel-by-pixel (Fig. 15). 3.7 Substrate–enzyme interaction and formation of the active complex give the complexity of the enzymatic reaction dynamics for conformation regulated enzymes The role of protein conformational changes in regulating the overall enzymatic activities has been extensively debated in recent years.19–23,26,30–32,35,39,129 These debates focus mainly on a few significant aspects: (1) There are two essentially orthogonal nuclear coordinates, one chemical reaction coordinate associated with enzymatic conversion of substrate to product (ES - EP) and the other conformational coordinate involved in substrate–enzyme complex formation and dissociation (E + S 2 ES) and product release (EP - E + P).9,24,29,36 Typically, along the chemical coordinates, the chemical reaction is a fast process (on the femtoseconds to submicroseconds time scales), presenting only minimum protein conformational changes.9,24,29,36 In contrast, along the conformational coordinates, the formation of the substrate–enzyme complex and the dissociation of the product–enzyme complex are significantly

1134 | Chem. Soc. Rev., 2014, 43, 1118--1143

Fig. 15 (upper panel) The TCAD map and cross correlation functions calculated from a simulated two-band fluctuation trajectory that consists of three sections: cross-correlated (I: 1–500 data points), non-correlated (II: 501–1000 data points) and anti-correlated (III: 1001–1500 data points). (A) 2D regional correlation analysis by TCAD mapping. The hot color represents the positive amplitude and the cold color represents the negative amplitude. (B) Correlation functions calculated from the three sections of correlated, non-correlated, and anti-correlated fluctuation data corresponding to sections I, II, III, and the whole data trajectory (I + II + III). It is evident that a conventional correlation calculation from the whole data trajectory gives no correlation amplitude, whereas the 2D regional correlation analysis gives definitive analysis of the correlation behavior for each specific fluctuation in the long trajectory. (lower panel) An example of 2D regional correlation analysis of single-molecule FRET fluctuation data. The experimental FRET two-band (D–A) fluorescence intensity fluctuation trajectory is measured from a D–A labeled kinase enzyme protein molecule showing a conformational change fluctuation in a buffer solution. (C) A TCAD map calculated from a two-band (D–A) FRET fluorescence fluctuation trajectory. The red and black trajectories are the donor and acceptor signals, respectively. (D) Cross correlation functions calculated from different sections of the trajectory. It is clear that the dynamics can be averaged out if only a whole-trajectory calculation is carried out. The anti-correlated FRET fluctuations can only dominate the fluorescence intensity trajectories in fraction of time periods but not all the time due to non-correlated and correlated thermal fluctuation background noises.39

slower (on the microseconds to seconds time scales), usually possessing large conformational motions that are observable by single-molecule FRET measurements.9,24,29,36 (2) Enzymatic reaction dynamics in the chemical coordinates shows Markovian or nonMarkovian, or even power-law dynamics,2–6,8,14,16,18,22,29,36,108,109,130, 131,133,136,137,139–142,146,158–161 whereas enzymatic dynamics along the conformational coordinates typically involves multiple-step conformational motions, such as bunching9,55,57 and even coherent conformational changes.44,47,55,132,135 (3) The rate-limiting step can be along either conformational or chemical reaction nuclear coordinates. In the measurements of single-molecule conformational dynamics and enzymatic turnovers of HPPK and T4 lysozyme,9 what we have probed is the dynamics in the conformational coordinate that is essential for forming the


View Article Online


Review Article

active complex for catalytic reactions. Due to the limited time resolution and signal-to-noise sensitivity, we are only able to probe the conformational dynamics along the conformational coordinates in substrate–enzyme complex formation and dissociation and product release, but not the dynamics along the chemical reaction coordinates. Presumably, the rate of a chemical reaction step (ES - EP) is significantly higher than the rate of conformational motion.39,132 There are potentially biological implications of the intrinsic bunched and even coherent conformational dynamics: for example, the formation of spatial and/or temporal biological complexity, structures, and function fluctuations in living cells. Compared to the above discussed time bunching effect on the hinge-bending conformational motion dynamics of T4 lysozyme under enzymatic reaction conditions,9,55,57,58 the coherence in the enzyme conformational state changes of the HPPK loop 2 conformational dynamics44,47 is a specific rate process of the bunched conformational motion dynamics, and both are originated from the intrinsic and convoluted multiple conformational intermediate states in each enzymatic reaction turnover cycle.

Chem Soc Rev

and inhomogeneous dynamics of enzyme–cell wall binding interactions, association and dissociation, and enzyme diffusion motions in the enzymatic reaction process. Overall, it is still a mystery how the T4 lysozyme efficiently hydrolyzes a cell wall, which is a covalently bonded polymer network with a heterogeneous structure and inhomogeneous electrostatic distribution. To directly probe the T4 lysozyme enzymatic reaction on a bacterial cell wall, we have developed and applied a combined single-molecule placement approach and spectroscopy analyses (Fig. 16).49 By placing a FRET donor–acceptor dye-labeled single T4 lysozyme molecule on a targeted bacterial cell wall by using a nano-liter hydrodynamic injection (Fig. 16A and B), we monitored single-molecule rotational motions during binding,

4. Towards interrogating enzymatic conformational dynamics in living cells and under active conformational manipulations Single-molecule spectroscopy has been demonstrated as a powerful technical approach to characterize fluctuating and inhomogeneous properties of enzymes, and it will be even more exciting to explore the hidden and non-detectable properties beyond the conventional observation-based single-molecule spectroscopy. One of the most desirable and promising developments is to study protein dynamics under active single-molecule manipulations: actively interactive manipulation and observation of single-molecule enzymes in action. Interestingly, recent development of this new frontier of single-molecule enzymology aims to demonstrate two major capabilities: (1) manipulating protein conformation coordinates that regulate the enzyme activities by using the atomic force microscope (AFM) tip to pull, hold, and periodically oscillate a specific residue site of the protein under a physiological condition; (2) explore and characterize the reactivity of the manipulated enzyme conformations at the single-molecule level with high spatial and temporal resolutions by using single-molecule spectroscopy imaging. In this section, we will focus our discussion on specifically two new types of singlemolecule spectroscopy approaches, as examples. 4.1 Placing single-molecule T4 lysozyme enzymes on a bacterial cell surface: toward studying single molecule enzymatic reaction dynamics in living cells T4 lysozyme can attach and bind to the cell wall and degrade the cell wall by catalyzing the hydrolysis of cell wall peptidoglycan. Much is still largely unknown about the complex mechanism


Fig. 16 Placing single-molecule T4 lysozyme enzymes on a bacterial cell surface. (A) Experiment setup. The E. coli cells were immobilized on a clean coverslip. The excitation laser was focused on the cell. A glass micropipette filled with enzyme solution was placed near the cell on the focal point by a micromanipulator. The solution in the micropipette was injected by a picoliter injector. We used a hydrodynamic nanoliter liquid injection technique using a micropipette. A picoliter injector (Harvard/Medical Systems PLI-100) was used to inject a controlled volume of solution to a cell wall substrate on a glass surface. The tip was placed 2–3 mm from the laser focal point where a cell wall piece was imaged. (B) A segment of a typical fluorescence time trajectory of injecting 108 M WT-Alexa 488 to a cell wall. Injection occurred every 4 seconds. We attribute the fluorescence intensity peaks to single molecules because they are quantized and drop to the background level in one step and because their intensity levels are similar to the intensity of immobilized single T4 lysozymes on a glass surface. Furthermore, the peaks are more likely to be observed immediately after injection as opposed to at random time points after injection. (inset) The fluorescence time trajectory of injecting 105 M Alexa 488 into the laser focal point for a duration of 20 ms. The counting dwell time of the trajectory was 10 ms. Panel B shows that the fluorescence intensity jumped to a higher level once a T4 lysozyme was delivered and bound to the cell wall after the injection pulse, and that the intensity dropped back to the background level after the molecule was photobleached or detached from the cell wall. In our single-molecule injection imaging experiments, autofluorescence from the cell was minimal because before the injection sequence began, the cell wall was pre-photobleached by laser with 100-times stronger power over a period of minutes. However, most molecules in the injection pulse flow away so that they would not be able to bind to or even collide with the cell wall. Typically, the ratio of the single-molecule placing is 1 out of 3–10 injection pulses.49

Chem. Soc. Rev., 2014, 43, 1118--1143 | 1135

View Article Online


Chem Soc Rev

attachment to, and dissociation from the cell wall by tracing single-molecule fluorescence intensity time trajectories and polarization. Probing the fluorescence polarization changes, we observed that the motions of the T4 lysozyme are associated with orientational rotations on the cell wall. This observation also suggests that the T4 lysozyme binding–unbinding motions on cell walls involve a complex mechanism beyond a single-step first-order rate process. By controlling the injection volume and concentration of the solution, we were able to deliver a single enzyme molecule to the cell wall resulting from an individual injection pulse (Fig. 16). Two types of T4 lysozyme bindings to cell walls are possible: (1) non-specific attachment and (2) chemical binding associated with the hydrolysis reaction. Events indicated by the fluorescence intensity dropping to background level are associated with T4 lysozyme diffusing away from the cell wall. When the T4 lysozyme attached to the cell wall, many enzymatic reaction turnovers likely occurred; in experiments, we have observed that the cell wall typically shrinks and eventually disappears from the imaging field of view.9,49 4.2 Manipulating protein conformations by single-molecule AFM-FRET nanoscopy Protein functions in enzymatic catalysis and protein–protein interactions are intrinsically associated with protein conformational fluctuations and folding-binding cooperative interactions.20,59,147,169 An enzyme can have different activities with different conformations,1–39,44–77,122,170,171 and conformational changes can significantly change the affinity and selectivity of protein interactions, which in turn often contribute to dramatic changes in protein functions.24,155 Thus, manipulating protein conformations can be effective for changing, enhancing, or even creating protein functions. It has been theoretically suggested that an oscillating force applied to an enzyme at a comparable frequency of enzymatic reaction turnover rate changes the enzymatic reaction activities due to force modification of the reaction pathway, potential surface, and enzymatic reaction intermediate state energy.121,172 In recent years, experimental works have demonstrated that an external mechanical force can change protein activities;82,92,95,103 accordingly, real-time measurements of protein conformational dynamics with a combined external force to manipulate and even control protein structures are a promising approach for protein structure– function studies. By combining atomic force microscopy and fluorescence resonance energy transfer spectroscopy (AFM-FRET), we have developed a single-molecule AFM-FRET nanoscopy approach capable of effectively pinpointing and mechanically manipulating a targeted dye-labeled single protein in a large sampling area, and simultaneously monitoring the conformational changes of the targeted protein by recording single-molecule FRET time trajectories. By analyzing time-resolved FRET trajectories and correlated AFM force pulling curves of the targeted single-molecule enzyme, we are able to observe the protein conformational changes of a specific coordination by AFM mechanic force pulling. Specifically, using our newly developed AFM-FRET nanoscopy

1136 | Chem. Soc. Rev., 2014, 43, 1118--1143

Review Article

Fig. 17 (A) Crystal structure of the Apo HPPK. The cyan spirals represent the a helices and the yellow arrows are the b strands. The loops are shown as red pipes. Amino acid residues 48 and 151 are labeled with FRET dye pair Cy3 and Cy5, respectively. The drawing is derived from the Protein Data Bank (1HKA). (B) Enzymatic reaction of HPPK-catalyzed pyrophosphoryl transfer. The dynamic processes comprise (1) the binding of two substrates (ATP and HP) to the enzyme (E) to form the enzyme–substrate complex (ES) and (2) the enzymatic turnover of the enzyme–substrate complex and the release of products (P).39

we were able to record and analyze single-molecule FRET time trajectories of a Cy3–Cy5 labeled kinase protein and correlated force spectroscopy on the same individual molecule. HPPK, an 18 kDa 158-residue monomeric protein (Fig. 17A), catalyzes the pyrophosphorylation reaction for the formation of 6-hydroxymethyl-7,8-dihydropterin pyrophosphate (HPPP) from 6-hydroxymethyl-7,8-dihydropterin (HP) reacting with adenosine-5 0 -triphosphate (ATP) (Fig. 17B), and leads to the biosynthesis of folates, a vitamin essential for life.44,55,57,173,174 HPPK consists of three flexible catalytic loops that are critical to the enzymatic reaction activity.44,55,57,173,174 Among the three catalytic loops, loops 2 and 3 undergo significant open–close conformational changes in each catalytic cycle, correlating with substrate (HP and ATP) binding and product (HPPP and AMP) release. It has been shown that the residues in loop 2 bind with one of the substrates, HP, while the residues in loop 3 interact with the substrate, ATP.44,55,57,173,174 To probe the single-molecule conformational change of protein under the AFM force pulling perturbation, the enzyme, HPPK, was labeled with Cy3–Cy5 at the amino acid residue 88 on loop 3 and residue 142 on the protein core close to the active site of the enzyme,37 respectively (Fig. 17). To apply mechanical force to perturb the conformation of the single enzyme molecule using the AFM tip, we coupled HPPK molecules between a glass coverslip and a ‘‘handle’’ function group (biotin and streptavidin) for the AFM tip through amine groups on the protein (Fig. 18). In our single-molecule AFM-FRET nanoscopy, time-resolved FRET trajectories and correlated AFM force pulling curves of the targeted single kinase enzyme are simultaneously recorded during the whole pulling approach-retract cycle. Fig. 18 presents


View Article Online


Review Article

Fig. 18 (A) Single-molecule AFM-FRET ultra nanoscopy; the zoomed panel in the left represents the schematic diagram of a FRET dye-pair (donor–acceptor: Cy3–Cy5) labeled HPPK molecule tethered between the surface of a glass coverslip and a handle (biotin group plus streptavidin), and another biotin group is modified on the AFM tip. (B) The correlated force curve; the curve shows the extension length of 24 nm within a period of 0.04 s. (C) A typical FRET time trajectory of the donor (green) and the acceptor (red) associated with one single-molecule AFMFRET force pulling event. Single-molecule fluorescence photon counting trajectories of the donor (Cy3, green) and the accepter (Cy5, red) are shown.

the typical data recorded from an effective AFM pulling event, showing the FRET donor–acceptor intensity trajectories, the correlated FRET efficiency trajectory, and the correlated AFM force curve. Besides correlated single-molecule AFM force spectroscopy and fluorescence FRET spectroscopy analysis, the AFM-FRET nanoscopy also has a high capability of conducting detailed force manipulation of enzyme conformations. The experimental results (Fig. 19), obtained in tris-buffer plus MgCl2 in the presence of an enzyme prohibitor (AMPCPP, a,b-methyleneadenosine 5 0 -triphosphate), show three typical force curves. We note that the results are not the eventual active manipulation of enzyme conformation and function yet, but rather show the capability of protein conformational manipulation in the combined AFM-FRET nanoscopy approach towards active manipulation of protein functions.45,46,82 Nevertheless, in Fig. 19, the force curves consist of saw tooth shaped peaks. These peaks are the results of unfolding of single HPPK molecules. The distances to rupture the protein from the force curves are 9 2 nm, 22 3 nm, and 46 7 nm, respectively (Fig. 19D). These results correspond to the possible unfolding configurations of the protein domains between amino acid residue 142 and amino acid residues 119, 85, and 23, respectively. In addition, these experimental results (9 2 nm, 22 3 nm, and 46 7 nm) are consistent with the theoretical results (8.7 nm, 21.7 nm, and 45.2 nm, respectively). As shown in Fig. 19B, DomB and DomC are unfolded with the rupture distance around 22 nm; this experimental result is also consistent with the theoretical value of 21.7 nm and the second peak (22 nm) of rupture distance


Chem Soc Rev

Fig. 19 (A–C) Three types of single-molecule force pulling curves of HPPK, as HPPK was chemically linked to a glass coverslip at residue 142, and AFM tip pulling occurs at the possible lysine residue sites 119, 85, and 23. In the insets above the force curves, three proposed domains are colored (green for DomA, purple for DomB and red for DomC) and depicted. (A) The unfolding force curve of DomC (red), which corresponds to the rupture distance of 9 nm. (B) The unfolding force curves of DomB (purple) and DomC (red), corresponding to the rupture distance of 22 nm. (C) The unfolding force curve of DoA, DoB and DoC, and the rupture distance is 45 nm. (D) Histogram of protein rupture distance distribution. The distribution of the rupture distances shows three peaks, at about 9 nm (DomC), 22 nm (DomB and DomC), and 45 nm (DomA, DomB and Dom C). (E) The structure of the HPPK mutant (the sites of lysines and cysteine are illustrated). Amino acid residue 142 was mutated to cysteine for specific site tethering of HPPK on the glass coverslip.45

distribution (Fig. 19D). In the third configuration (23, 142), as shown in Fig. 19C, three proposed domains (DomA, DomB and DomC) between site 23 and site 142 are all unfolded, thus giving a larger rupture distance compared to the unfolding of only DomC (Fig. 19A) or DomB and DomC (Fig. 19B). Furthermore, the experimental value (45 nm), the expected theoretical value (45.2 nm) and the third peak (45 nm) in the distribution of rupture distance are all almost identical to each other. Overall, using this approach, we are able (1) to locate an individual Cy3–Cy5 labeled enzyme molecule with pinpoint nanoscale precision; (2) to mechanically manipulate the enzyme conformations by force pulling and unfolding the target single enzyme molecule; and (3) to simultaneously probe the protein conformational changes associated with its enzymatic activity by single-molecule FRET spectroscopy measurements during the AFM pulling manipulation. Our AFM-FRET nanoscopy presents a significant advancement compared to current reported techniques80,82,94,95,103 in terms of conducting simultaneous single-molecule force manipulation and FRET measurements probing the corresponding conformational changes of a single targeted enzyme molecule, which is particularly powerful for studying enzyme function-conformation mechanisms and relationships between function and conformations.

5. Concluding remarks Studying enzymatic dynamics and the associated active-site conformational dynamics calls for a systematic research approach, which relies on the molecular-level understanding of molecular

Chem. Soc. Rev., 2014, 43, 1118--1143 | 1137

View Article Online


Chem Soc Rev

interactions, local environment fluctuations, energy landscapes, and various rate processes along various nuclear coordinates. The complexity and inhomogeneity of enzymatic dynamics require that the knowledge has to be obtained from the single-molecule analyses at spatial and temporal resolutions, beyond the conventional solution-phase ensemble-averaged analyses. Single-molecule spectroscopy is particularly powerful for identifying and characterizing spatially and temporally inhomogeneous and complex enzymatic active-site conformational dynamics. In light of the rapid developments, the most current and future developments and applications of the single-molecule spectroscopy will definitely be benefited by further advanced single-molecule approaches: spectroscopy combined experimental multiple-parameters with high chemical selectivity and wide temporal and spatial resolutions; computational molecular dynamics simulation analysis, based on static protein structures and correlated experimental singlemolecule trajectory analyses; and theoretical modeling with experimentally accessible physical parameters. It is the active complex formation processes (E + S - ES - ES*) that define the enzymatic reaction potential surfaces and contribute to the complexity and inhomogeneity of the enzymatic reactions. Understanding enzymatic reaction conformational dynamics is intimately related to single-molecule studies of biomolecular interactions for the precise reason that the formation of an enzyme–substrate active complex in biomolecular interactions. In recent years, the mechanisms of protein conformation selection and induced conformational changes have been extensively explored, and it is anticipated that more single-molecule protein–protein interaction studies will also contribute to our fundamental understanding of enzymatic reaction dynamics and mechanisms. We now have an unprecedented opportunity to apply many of the new technical and scientific molecular science approaches to studying the enzymatic dynamics and associated protein dynamics; for example, the next generation of single-molecule spectroscopy will be not only capable of probing complex singlemolecule protein dynamics but also capable of manipulating protein conformational dynamics and control protein activities. The new single-molecule spectroscopy will be powerful in exploring unprecedented protein properties, protein dynamics, and protein function energetic landscapes. Ultimately, our fundamental understanding of the enzymology, especially of the enzymatic dynamics, will facilitate new medicines, human health development, and new concept of the nature of diseases, health care, and the nature of life.

Acknowledgements The author gratefully acknowledges Dehong Hu, Yu Chen, Erich Vorpagel, Yufan He, Yuanmin Wang, Xuefei Wang, Saptarshi Mukherjee, Desheng Zheng, Maolin Lu, Jin Cao, and Leonora Kaldaras for their crucial contributions to the work discussed here; Brian Matthews and Honggao Yang for providing the enzyme samples of T4 lysozyme and HPPK and for the stimulating discussions. The author acknowledges the support to his program from the National Institute of General Medicine

1138 | Chem. Soc. Rev., 2014, 43, 1118--1143

Review Article

Sciences (NIGMS) of NIH, Bowling Green State University, Pacific Northwest National Laboratory, The super computer facility grant of the Environmental Molecular Sciences Laboratory of PNNL, and Ohio Eminent Scholar Endowment. Part of the text has appeared in the author’s previous publications referenced.

References 1 D. B. Craig, E. A. Arriaga, J. C. Y. Wong, H. Lu and N. J. Dovichi, Studies on single alkaline phosphatase molecules: reaction rate and activation energy of a reaction catalyzed by a single molecule and the effect of thermal denaturation – the death of an enzyme, J. Am. Chem. Soc., 1996, 118(22), 5245–5253. 2 L. Edman and R. Rigler, Memory landscapes of singleenzyme molecules, Proc. Natl. Acad. Sci. U. S. A., 2000, 97(15), 8266–8271. 3 B. P. English, et al., Ever-fluctuating single enzyme molecules: Michaelis–Menten equation revisited, Nat. Chem. Biol., 2006, 2(2), 87–94. 4 H. P. Lerch, R. Rigler and A. S. Mikhailov, Functional conformational motions in the turnover cycle of cholesterol oxidase, Proc. Natl. Acad. Sci. U. S. A., 2005, 102(31), 10807–10812. 5 H. P. Lu, L. Y. Xun and X. S. Xie, Single-molecule enzymatic dynamics, Science, 1998, 282(5395), 1877–1882. 6 X. S. Xie and H. P. Lu, Single-molecule enzymology, J. Biol. Chem., 1999, 274(23), 15967–15970. 7 Q. F. Xue and E. S. Yeung, Differences in The ChemicatReactivity of Individual Molecules of an Enzyme, Nature, 1995, 373(6516), 681–683. 8 N. Agmon, Conformational cycle of a single working enzyme, J. Phys. Chem. B, 2000, 104(32), 7830–7834. 9 Y. Chen, D. H. Hu, E. R. Vorpagel and H. P. Lu, Probing single-molecule T4 lysozyme conformational dynamics by intramolecular fluorescence energy transfer, J. Phys. Chem. B, 2003, 107(31), 7947–7956. 10 Y. Ishii and T. Yanagida, Single Molecule Detectionin Life Science, Single Mol., 2000, 1, 5–16. 11 H. P. Lu, L. M. Iakoucheva and E. J. Ackerman, Singlemolecule DNA damage recognition: DNA-protein proteinprotein interaction dynamics, Biophys. J., 2000, 78(1), 262A. 12 H. P. Lu, L. M. Iakoucheva and E. J. Ackerman, Singlemolecule conformational dynamics of fluctuating noncovalent DNA–Protein interactions a in DNA damage recognition, J. Am. Chem. Soc., 2001, 123(37), 9184–9185. 13 W. E. Moerner and M. Orrit, Illuminating single molecules in condensed matter, Science, 1999, 283(5408), 1670–1676. 14 G. K. Schenter, H. P. Lu and X. S. Xie, Statistical analyses and theoretical models of single-molecule enzymatic dynamics, J. Phys. Chem. A, 1999, 103(49), 10477–10488. 15 A. M. van Oijen, et al., Single-molecule kinetics of lambda exonuclease reveal base dependence and dynamic disorder, Science, 2003, 301(5637), 1235–1238.


View Article Online


Review Article

16 J. Wang and P. Wolynes, Intermittency of Single-Molecule Reaction Dynamics in Fluctuating Environments, Phys. Rev. Lett., 1995, 74(21), 4317–4320. 17 X. S. Xie and J. K. Trautman, Optical studies of single molecules at room temperature, Annu. Rev. Phys. Chem., 1998, 49, 441–480. 18 R. Zwanzig, Rate-Processes with Dynamic Disorder, Acc. Chem. Res., 1990, 23(5), 148–152. 19 N. M. Antikainen, R. D. Smiley, S. J. Benkovic and G. G. Hammes, Conformation coupled enzyme catalysis: singlemolecule and transient kinetics investigation of dihydrofolate reductase, Biochemistry, 2005, 44(51), 16835–16843. 20 E. Z. Eisenmesser, D. A. Bosco, M. Akke and D. Kern, Enzyme dynamics during catalysis, Science, 2002, 295(5559), 1520–1523. 21 E. Z. Eisenmesser, et al., Intrinsic dynamics of an enzyme underlies catalysis, Nature, 2005, 438(7064), 117–121. 22 J. Gao, et al., Mechanisms and free energies of enzymatic reactions, Chem. Rev., 2006, 106(8), 3188–3209. 23 M. Garcia-Viloca, J. Gao, M. Karplus and D. G. Truhlar, How enzymes work: analysis by modern rate theory and computer simulations, Science, 2004, 303(5655), 186–195. 24 Single Molecule Spectroscopy, in Chemistry, Physics and Biology (Nobel Symposium), ed. A. Graslund, R. Rigler and J. Widengren, Springer, 2010. 25 T. J. Ha, et al., Single-molecule fluorescence spectroscopy of enzyme conformational dynamics and cleavage mechanism, Proc. Natl. Acad. Sci. U. S. A., 1999, 96(3), 893–898. 26 K. A. Henzler-Wildman, et al., Intrinsic motions along an enzymatic reaction trajectory, Nature, 2007, 450(7171), 838–844. 27 Y. Ishii, K. Kitamura, H. Tanaka and T. Yanagida, Molecular motors and single-molecule enzymology, Biophotonics, Pt B, 2003, 361, 228–245. 28 S. X. Liu, G. Bokinsky, N. G. Walter and X. W. Zhuang, Dissecting the multistep reaction pathway of an RNA enzyme by single-molecule kinetic ‘‘fingerprinting’’, Proc. Natl. Acad. Sci. U. S. A., 2007, 104(31), 12634–12639. 29 H. P. Lu, Probing single-molecule protein conformational dynamics, Acc. Chem. Res., 2005, 38(7), 557–565. 30 W. Min, et al., Fluctuating enzymes: ressons from singlemolecule studies, Acc. Chem. Res., 2005, 38(12), 923–931. 31 W. Min, X. S. Xie and B. Bagchi, Role of conformational dynamics in kinetics of an enzymatic cycle in a nonequilibrium steady state, J. Chem. Phys., 2009, 131(6), 065104. 32 T. Mittag, L. E. Kay and J. D. Forman-Kay, Protein dynamics and conformational disorder in molecular recognition, J. Mol. Recognit., 2009, 23(2), 105–116. 33 W. E. Moerner, New directions in single-molecule imaging and analysis, Proc. Natl. Acad. Sci. U. S. A., 2007, 104(31), 12596–12602. 34 M. Orrit, The motions of an enzyme soloist, Science, 2003, 302(5643), 239–240. 35 A. V. Pisliakov, J. Cao, S. C. L. Kamerlin and A. Warshel, Enzyme millisecond conformational dynamics do not catalyze the chemical step, Proc. Natl. Acad. Sci. U. S. A., 2009, 106(41), 17359–17364.


Chem Soc Rev

36 Single Molecule Spectroscopy: Nobel Conference Lectures, ed. R. Rigler, M. Orrit and T. Basche, Springer-Verlag, Berlin, 2003. 37 G. Seisenberger, et al., Real-time single-molecule imaging of the infection pathway of an adeno-associated virus, Science, 2001, 294(5548), 1929–1932. 38 K. Visscher, M. J. Schnitzer and S. M. Block, Single kinesin molecules studied with a molecular force clamp, Nature, 1999, 400(6740), 184–189. 39 P. C. Whitford, J. N. Onuchic and P. G. Wolynes, Energy landscape along an enzymatic reaction trajectory: hinges or cracks?, HFSP J., 2008, 2(2), 61–64. 40 P. W. Fenimore, H. Frauenfelder, B. H. McMahon and R. D. Young, Bulk-solvent and hydration-shell fluctuations, similar to alpha- and beta-fluctuations in glasses, control protein motions and functions, Proc. Natl. Acad. Sci. U. S. A., 2004, 101(40), 14408–14413. 41 P. W. Fenimore, H. Frauenfelder, B. H. McMahon and R. D. Young, Proteins are paradigms of stochastic complexity, Phys. A, 2005, 351(1), 1–13. 42 H. Frauenfelder, S. G. Sligar and P. G. Wolynes, The energy landscapes and motions of proteins, Science, 1991, 254(5038), 1598–1603. 43 H. Frauenfelder, P. G. Wolynes and R. H. Austin, Biological physics, Rev. Mod. Phys., 1999, 71(2), S419–S430. 44 Y. He, et al., Probing Single-Molecule Enzyme Active-Site Conformational State Intermittent Coherence, J. Am. Chem. Soc., 2011, 133(36), 14389–14395. 45 Y. He, M. Lu, J. Cao and H. P. Lu, Manipulating Protein Conformations by Single-Molecule AFM-FRET Nanoscopy, ACS Nano, 2012, 6(2), 1221–1229. 46 Y. He, M. Lu and H. P. Lu, Single-molecule photon stamping FRET spectroscopy study of enzymatic conformational dynamics, Phys. Chem. Chem. Phys., 2013, 15(3), 770–775. 47 H. P. Lu, Enzymes in Coherent Motion, Science, 2012, 335(6066), 300–301. 48 M. Margittai, et al., Single-molecule fluorescence resonance energy transfer reveals a dynamic equilibrium between closed and open conformations of syntaxin 1, Proc. Natl. Acad. Sci. U. S. A., 2003, 100(26), 15516–15521. 49 D. H. Hu and H. P. Lu, Placing single-molecule T4 lysozyme enzymes on a bacterial cell surface: toward probing single-molecule enzymatic reaction in living cells, Biophys. J., 2004, 87(1), 656–661. 50 R. C. Liu, D. H. Hu, X. Tan and H. P. Lu, Revealing twostate protein-protein interactions of calmodulin by singlemolecule spectroscopy, J. Am. Chem. Soc., 2006, 128(31), 10034–10042. 51 H. P. Lu, Single-molecule spectroscopy studies of conformational change dynamics in enzymatic reactions, Curr. Pharm. Biotechnol., 2004, 5(3), 261–269. 52 H. P. Lu, Combined Single-Molecule Electrical Recording and Single-Molecule Spectroscopy Studies of Ion Channel Conformational Dynamics, in Methods in Nano Cell Biology, ed. B. P. Jena, 2008, vol. 90, pp. 435–451.

Chem. Soc. Rev., 2014, 43, 1118--1143 | 1139

View Article Online


Chem Soc Rev

53 H. P. Lu, Single-Molecule Protein Interaction Conformational Dynamics, Curr. Pharm. Biotechnol., 2009, 10(5), 522–531. 54 H. P. Lu, Single-Molecule Protein Conformational Dynamics in Enzymatic Reactions, in Single Molecule Spectroscopy in Chemistry, Physics and Biology, ed. A. Graslund, R. Rigler and J. Widengren, 2010, vol. 96, pp. 471–494. 55 H. P. Lu, Revealing time bunching effect in singlemolecule enzyme conformational dynamics, Phys. Chem. Chem. Phys., 2011, 13(15), 6734–6749. 56 Q. Lu, H. P. Lu and J. Wang, Exploring the mechanism of flexible biomolecular recognition with single molecule dynamics, Phys. Rev. Lett., 2007, 98(12), 128105. 57 Y. M. Wang and H. P. Lu, Bunching Effect in SingleMolecule T4 Lysozyme Nonequilibrium Conformational Dynamics under Enzymatic Reactions, J. Phys. Chem. B, 2010, 114(19), 6669–6674. 58 H. P. Lu, Single-molecule study of protein-protein and protein-DNA interaction dynamics, in Methods in Molecular Biology, ed. G. U. Nienhaus, 2005, vol. 305, pp. 385–413. 59 X. Tan, et al., Single-molecule study of protein-protein interaction dynamics in a cell signaling system, J. Phys. Chem. B, 2004, 108(2), 737–744. 60 J. Wang, Q. Lu and H. P. Lu, Single-molecule dynamics reveals cooperative binding-folding in protein recognition, PLoS Comput. Biol., 2006, 2(7), 842–852. 61 X. Wang and H. P. Lu, 2D Regional Correlation Analysis of Single-Molecule Time Trajectories, J. Phys. Chem. B, 2008, 112(47), 14920–14926. 62 S. Kim, et al., Probing Allostery Through DNA, Science, 2013, 339(6121), 816–819. 63 W. Min, L. Jiang and X. S. Xie, Complex Kinetics of Fluctuating Enzymes: Phase Diagram Characterization of a Minimal Kinetic Scheme, Chem.–Asian J., 2010, 5(5), 1129–1138. 64 W. Min, X. S. Xie and B. Bagchi, Two-dimensional reaction free energy surfaces of catalytic reaction: effects of protein conformational dynamics on enzyme catalysis, J. Phys. Chem. B, 2008, 112(2), 454–466. 65 X. S. Xie, Enzymology and Life at the Single Molecule Level, in Single Molecule Spectroscopy in Chemistry, Physics and Biology, ed. A. Graslund, R. Rigler and J. Widengren, 2010, vol. 96, pp. 435–448. 66 Y. Choi, et al., Single-Molecule Lysozyme Dynamics Monitored by an Electronic Circuit, Science, 2012, 335(6066), 319–324. 67 Y. Choi, et al., Single-Molecule Dynamics of Lysozyme Processing Distinguishes Linear and Cross-Linked Peptidoglycan Substrates, J. Am. Chem. Soc., 2012, 134(4), 2032–2035. 68 Y. Choi, et al., Dissecting Single-Molecule Signal Transduction in Carbon Nanotube Circuits with Protein Engineering, Nano Lett., 2013, 13(2), 625–631. 69 P. C. Sims, et al., Electronic Measurements of SingleMolecule Catalysis by cAMP-Dependent Protein Kinase A, J. Am. Chem. Soc., 2013, 135(21), 7861–7868.

1140 | Chem. Soc. Rev., 2014, 43, 1118--1143

Review Article

70 J. A. Hanson, et al., Illuminating the mechanistic roles of enzyme conformational dynamics, Proc. Natl. Acad. Sci. U. S. A., 2007, 104(46), 18055–18060. 71 W. P. Jencks, Binding-Energy, Specificity, and Enzymic Catalysis-Circe Effect, Adv. Enzymol. Relat. Areas Mol. Biol., 1975, 43, 219–410. 72 P. Maragakis and M. Karplus, Large amplitude conformational change in proteins explored with a plastic network model: adenylate kinase, J. Mol. Biol., 2005, 352(4), 807–822. 73 O. Miyashita, J. N. Onuchic and P. G. Wolynes, Nonlinear elasticity, proteinquakes, and the energy landscapes of functional transitions in proteins, Proc. Natl. Acad. Sci. U. S. A., 2003, 100(22), 12570–12575. 74 E. M. Puchner and H. E. Gaub, Single-Molecule Mechanoenzymatics, in Annu Rev Biophys, ed. D. C. Rees, 2012, vol. 41, pp. 497–518. 75 C. Vonrhein, G. J. Schlauderer and G. E. Schulz, Movie of The Structural-Changes During a catalytic Cycle of Nucleoside Monophosphate Kinases, Structure, 1995, 3(5), 483–490. 76 H. G. Yan and M. D. Tsai, Nucleoside monophosphate kinases: structure, mechanism, and substrate specificity, Adv. Enzymol., 1999, 73, 103–134. 77 X. W. Zhuang, et al., A single-molecule study of RNA catalysis and folding, Science, 2000, 288(5473), 2048–2051. 78 W. A. Catterall, Structure and Function of Voltage-Sensitive Ion Channels, Science, 1988, 242(4875), 50–61. 79 A. A. Deniz, et al., Single-molecule protein folding: diffusion fluorescence resonance energy transfer studies of the denaturation of chymotrypsin inhibitor 2, Proc. Natl. Acad. Sci. U. S. A., 2000, 97(10), 5179–5184. 80 A. Gaiduk, et al., Fluorescence detection with high time resolution: from optical microscopy to simultaneous force and fluorescence spectroscopy, Microsc. Res. Tech., 2007, 70(5), 433–441. 81 I. V. Gopich and A. Szabo, Theory of the energy transfer efficiency and fluorescence lifetime distribution in singlemolecule FRET, Proc. Natl. Acad. Sci. U. S. A., 2012, 109(20), 7747–7752. 82 H. Gumpp, et al., Triggering Enzymatic Activity with Force, Nano Lett., 2009, 9(9), 3290–3295. 83 T. Ha, et al., Probing the interaction between two single molecules: fluorescence resonance energy transfer between a single donor and a single acceptor, Proc. Natl. Acad. Sci. U. S. A., 1996, 93(13), 6264–6268. 84 E. Haas, The study of protein folding and dynamics by determination of intramolecular distance distributions and their fluctuations using ensemble and single-molecule FRET measurements, ChemPhysChem, 2005, 6(5), 858–870. 85 G. Harms, G. Orr and H. P. Lu, Probing ion channel conformational dynamics using simultaneous singlemolecule ultrafast spectroscopy and patch-clamp electric recording, Appl. Phys. Lett., 2004, 84(10), 1792–1794. 86 G. S. Harms, et al., Probing conformational changes of gramicidin ion channels by single-molecule patch-clamp


View Article Online

Review Article


87

88

89 90

91

92

93

94

95

96

97 98

99 100 101

102 103

fluorescence microscopy, Biophys. J., 2003, 85(3), 1826–1838. B. Hille, Ionic Channels of Excitable Membranes, Sinauer Associates, Inc., Sunderland, Massachusetts USA., 2nd edn, 1992. T. Ide, M. Hirano and Y. Takeuchi, Ion Channels, in Single Molecule Dynamics in Life Science, ed. T. Yanagida and Y. Ishii, Wiley-VCH, 2009, pp. 87–97. M. O. Jensen, et al., Mechanism of Voltage Gating in Potassium Channels, Science, 2012, 336(6078), 229–233. Y. W. Jia, et al., Folding dynamics of single GCN4 peptides by fluorescence resonant energy transfer confocal microscopy, Chem. Phys., 1999, 247(1), 69–83. J. P. Junker and M. Rief, Single-molecule force spectroscopy distinguishes target binding modes of calmodulin, Proc. Natl. Acad. Sci. U. S. A., 2009, 106(34), 14361–14366. J. P. Junker, F. Ziegler and M. Rief, Ligand-Dependent Equilibrium Fluctuations of Single Calmodulin Molecules, Science, 2009, 323(5914), 633–637. S. Kalinin, A. Valeri, M. Antonik, S. Felekyan and C. A. M. Seidel, Detection of Structural Dynamics by FRET: A Photon Distribution and Fluorescence Lifetime Analysis of Systems with Multiple States, J. Phys. Chem. B, 2010, 114(23), 7983–7995. M. S. Z. Kellermayer, et al., Spatially and temporally synchronized atomic force and total internal reflection fluorescence microscopy for imaging and manipulating cells and biomolecules, Biophys. J., 2006, 91(7), 2665–2677. T. Kodama, H. Ohtani, H. Arakawa and A. Ikai, Mechanical perturbation-induced fluorescence change of green fluorescent protein, Appl. Phys. Lett., 2005, 86(4), 043901. R. Kuroki, L. H. Weaver and B. W. Matthews, A Covalent Enzyme-Substrate Intermediate with Saccharide Distortion in a Mutant T4 Lysozyme, Science, 1993, 262(5142), 2030–2033. J. R. Lakowicz, Principles of fluorescence spectroscopy, Springer, 3rd edn, 2006. Z. K. Majumdar, R. Hickerson, H. F. Noller and R. M. Clegg, Measurements of internal distance changes of the 30 S ribosome using FRET with multiple donoracceptor pairs: quantitative spectroscopic methods, J. Mol. Biol., 2005, 351(5), 1123–1145. B. W. Matthews, Study on Protein Stability with T4 Lysozyme, Adv. Protein. Chem., 1995, 46, 249–278. J. A. McCammon and S. H. Northrup, Gated Binding of Ligands to Proteins, Nature, 1981, 293(5830), 316–317. P. J. Rothwell, et al., Multiparameter single-molecule fluorescence spectroscopy reveals heterogeneity of HIV-1 reverse transcriptase: primer/template complexes, Proc. Natl. Acad. Sci. U. S. A., 2003, 100(4), 1655–1660. R. Roy, S. Hohng and T. Ha, A practical guide to singlemolecule FRET, Nat. Methods, 2008, 5(6), 507–516. A. Sarkar, R. B. Robertson and J. M. Fernandez, Simultaneous atomic force microscope and fluorescence measurements of protein unfolding using a calibrated evanescent wave, Proc. Natl. Acad. Sci. U. S. A., 2004, 101(35), 12882–12886.


Chem Soc Rev

104 B. Schuler, E. A. Lipman and W. A. Eaton, Probing the freeenergy surface for protein folding with single-molecule fluorescence spectroscopy, Nature, 2002, 419(6908), 743–747. 105 Single-Molecule Techniques: A Laboratory Manual, ed. P. R. Selvin and T. Ha, Cold Spring Harbor Laboratory Press, New York, 2008. 106 D. S. Talaga, et al., Dynamics and folding of single twostranded coiled-coil peptides studied by fluorescent energy transfer confocal microscopy, Proc. Natl. Acad. Sci. U. S. A., 2000, 97(24), 13021–13026. 107 J. Widengren, et al., Single-molecule detection and identification of multiple species by multiparameter fluorescence detection, Anal. Chem., 2006, 78(6), 2039–2050. 108 H. Yang, et al., Protein conformational dynamics probed by single-molecule electron transfer, Science, 2003, 302(5643), 262–266. 109 H. Yang and X. S. Xie, Probing single-molecule dynamics photon by photon, J. Chem. Phys., 2002, 117(24), 10965–10979. 110 P. V. Cornish and T. Ha, A survey of single-molecule techniques in chemical biology, ACS Chem. Biol., 2007, 2(1), 53–61. 111 C. Joo, H. Balci, Y. Ishitsuka, C. Buranachai and T. Ha, Advances in single-molecule fluorescence methods for molecular biology, Annu. Rev. Biochem., 2008, 77, 51–76. 112 I. Rasnik, S. A. McKinney and T. Ha, Surfaces and orientations: much to FRET about?, Acc. Chem. Res., 2005, 38(7), 542–548. 113 M. Sorokina, H.-R. Koh, S. S. Patel and T. Ha, Fluorescent Lifetime Trajectories of a Single Fluorophore Reveal Reaction Intermediates During Transcription Initiation, J. Am. Chem. Soc., 2009, 131(28), 9630–9631. 114 R. Zhou, M. Schlierf and T. Ha, Force Fluorescence Spectroscopy at The Single-Molecule Level, in Methods in Enzymology, Vol. 475: Single Molecule Tools, Pt B: SuperResolution, Particle Tracking, Multiparameter, and Force Based Methods, ed. N. G. Walter, 2010, vol. 475, pp. 405–426. 115 D. D. Boehr, D. McElheny, H. J. Dyson and P. E. Wright, The dynamic energy landscape of dihydrofolate reductase catalysis, Science, 2006, 313(5793), 1638–1642. 116 S. L. Chang and N. Tjandra, Temperature dependence of protein backbone motion from carbonyl C-13 and amide N-15 NMR relaxation, J. Magn. Reson., 2005, 174(1), 43–53. 117 K. A. Henzler-Wildman, et al., A hierarchy of timescales in protein dynamics is linked to enzyme catalysis, Nature, 2007, 450(7171), 913–916. 118 J. C. Williams and A. E. McDermott, Dynamics of The Flexible Loop of Triosephosphate Isomerase -The Loop Motion is not Ligand-Gated, Biochemistry, 1995, 34(26), 8309–8319. 119 A. V. Zhuravleva, et al., Gated electron transfers and electron pathways in azurin: A NMR dynamic study at multiple fields and temperatures, J. Mol. Biol., 2004, 342(5), 1599–1611.

Chem. Soc. Rev., 2014, 43, 1118--1143 | 1141

View Article Online


Chem Soc Rev

120 D. D. Boehr, H. J. Dyson and P. E. Wright, An NMR perspective on enzyme dynamics, Chem. Rev., 2006, 106(8), 3055–3079. 121 M. A. Lomholt, M. Urbakh, R. Metzler and J. Klafter, Manipulating single enzymes by an external harmonic force, Phys. Rev. Lett., 2007, 98(16), 168302. 122 A. K. Mittermaier and L. E. Kay, Observing biological dynamics at atomic resolution using NMR, Trends Biochem. Sci., 2009, 34(12), 601–611. 123 A. P. Wiita, R. K. Ainavarapu, H. H. Huang and J. M. Fernandez, Force-dependent chemical kinetics of disulfide bond reduction observed with single-molecule techniques, Proc. Natl. Acad. Sci. U. S. A., 2006, 103(19), 7222–7227. 124 A. P. Wiita, et al., Probing the chemistry of thioredoxin catalysis with force, Nature, 2007, 450(7166), 124–127. 125 J. S. Biteen, et al., Super-resolution imaging in live Caulobacter crescentus cells using photoswitchable EYFP, Nat. Methods, 2008, 5(11), 947–949. 126 A. E. Cohen and W. E. Moerner, Suppressing Brownian motion of individual biomolecules in solution, Proc. Natl. Acad. Sci. U. S. A., 2006, 103(12), 4362–4365. 127 W. E. Moerner, Microscopy beyond the diffraction limit using actively controlled single molecules, J. Microsc., 2012, 246(3), 213–220. 128 M. A. Thompson, J. M. Casolari, M. Badieirostami, P. O. Brown and W. E. Moerner, Three-dimensional tracking of single mRNA particles in Saccharomyces cerevisiae using a double-helix point spread function, Proc. Natl. Acad. Sci. U. S. A., 2010, 107(42), 17864–17871. 129 Q. Wang, R. H. Goldsmith, Y. Jiang, S. D. Bockenhauer and W. E. Moerner, Probing Single Biomolecules in Solution Using the Anti-Brownian Electrokinetic (ABEL) Trap, Acc. Chem. Res., 2012, 45(11), 1955–1964. 130 Y. He and E. Barkai, Super- and sub-Poissonian photon statistics for single molecule spectroscopy, J. Chem. Phys., 2005, 122(18), 184703. 131 Q. Lu and J. Wang, Single molecule conformational dynamics of adenylate kinase: Energy landscape, structural correlations, and transition state ensembles, J. Am. Chem. Soc., 2008, 130(14), 4772–4783. 132 H. Qian, Open-system nonequilibrium steady state: statistical thermodynamics, fluctuations, and chemical oscillations, J. Phys. Chem. B, 2006, 110(31), 15063–15074. 133 H. Qian and E. L. Elson, Single-molecule enzymology: stochastic Michaelis–Menten kinetics, Biophys. Chem., 2002, 101, 565–576. 134 K. Velonia, et al., Single-enzyme kinetics of CALB-catalyzed hydrolysis, Angew. Chem., Int. Ed., 2005, 44(4), 560–564. 135 M. O. Vlad, F. Moran, F. W. Schneider and J. Ross, Memory effects and oscillations in single-molecule kinetics, Proc. Natl. Acad. Sci. U. S. A., 2002, 99(20), 12548–12555. 136 S. L. Yang and J. S. Cao, Direct measurements of memory effects in single-molecule kinetics, J. Chem. Phys., 2002, 117(24), 10996–11009. 137 Y. J. Zheng and F. L. H. Brown, Single molecule photon emission statistics for non-Markovian blinking models, J. Chem. Phys., 2004, 121(7), 3238–3252.

1142 | Chem. Soc. Rev., 2014, 43, 1118--1143

Review Article

138 S. Kim, C. M. Schroeder and X. S. Xie, Single-Molecule Study of DNA Polymerization Activity of HIV-1 Reverse Transcriptase on DNA Templates, J. Mol. Biol., 2010, 395(5), 995–1006. 139 E. Barkai, Y. Jung and R. Silbey, Time-dependent fluctuations in single molecule spectroscopy: a generalized Wiener–Khintchine approach, Phys. Rev. Lett., 2001, 87(20), 207403. 140 E. Barkai, R. Silbey and G. Zumofen, Transition from simple to complex behavior of single molecule line shapes in disordered condensed phase, J. Chem. Phys., 2000, 113(14), 5853–5867. 141 D. D. Boehr and P. E. Wright, How do proteins interact?, Science, 2008, 320(5882), 1429–1430. 142 F. L. H. Brown, Single-molecule kinetics with timedependent rates: a generating function approach, Phys. Rev. Lett., 2003, 90(2), 028302. 143 Q. Cui and M. Karplus, Catalysis and specificity in enzymes: a study of triosephosphate isomerase and comparison with methyl glyoxal synthase, Adv. Protein Chem., 2003, 66, 315–372. 144 A. G. Palmer, NMR characterization of the dynamics of biomacromolecules, Chem. Rev., 2004, 104(8), 3623–3640. 145 M. Wolf-Watz, et al., Linkage between dynamics and catalysis in a thermophilic-mesophilic enzyme pair, Nat. Struct. Mol. Biol., 2004, 11(10), 945–949. 146 O. Flomenbom, et al., Stretched exponential decay and correlations in the catalytic activity of fluctuating single lipase molecules, Proc. Natl. Acad. Sci. U. S. A., 2005, 102(7), 2368–2372. 147 X. Tan, D. H. Hu, T. C. Squier and H. P. Lu, Probing nanosecond protein motions of calmodulin by singlemolecule fluorescence anisotropy, Appl. Phys. Lett., 2004, 85(12), 2420–2422. 148 S. R. Adams, et al., New biarsenical ligands and tetracysteine motifs for protein labeling in vitro and in vivo: synthesis and biological applications, J. Am. Chem. Soc., 2002, 124(21), 6063–6076. 149 B. A. Griffin, S. R. Adams and R. Y. Tsien, Specific covalent labeling of recombinant protein molecules inside live cells, Science, 1998, 281(5374), 269–272. 150 Y. F. Dufrene, et al., Five challenges to bringing singlemolecule force spectroscopy into living cells, Nat. Methods, 2011, 8(2), 123–127. 151 S. F. Heucke, et al., Nanoapertures for AFM-based singlemolecule force spectroscopy, Int. J. Nanotechnol., 2013, 10(5–7), 607–619. 152 E. M. Puchner and H. E. Gaub, Mechanoenzymatics and Nanoassembly of Single Molecules, in Single Molecule Spectroscopy in Chemistry, Physics and Biology, ed. A. Graslund, R. Rigler and J. Widengren, 2010, vol. 96, pp. 289–303. 153 M. Strackharn, D. A. Pippig, P. Meyer, S. W. Stahl and H. E. Gaub, Nanoscale Arrangement of Proteins by SingleMolecule Cut-and-Paste, J. Am. Chem. Soc., 2012, 134(37), 15193–15196.


View Article Online


Review Article

154 R. Perez-Jimenez, et al., Single-molecule paleoenzymology probes the chemistry of resurrected enzymes, Nat. Struct. Mol. Biol., 2011, 18(5), 592–596. 155 D. S. Zheng, L. Kaldaras and H. P. Lu, Total internal reflection fluorescence microscopy imaging-guided confocal single-molecule fluorescence spectroscopy, Rev. Sci. Instrum., 2012, 83(1), 013110. 156 G. E. Arnold and R. L. Ornstein, Protein hinge bending as seen in molecular dynamics simulations of native and M6I mutant T4 lysozymes, Biopolymers, 1997, 41(5), 533–544. 157 A. Tsugita, M. Inouye, E. Terzaghi and G. Streisinger, Purification of Bacteriophage T4 Lysozyme, J. Biol. Chem., 1968, 243(2), 391–397. 158 J. Cao, Michaelis–Menten Equation and Detailed Balance in Enzymatic Networks, J. Phys. Chem. B, 2011, 115(18), 5493–5498. 159 J. S. Cao, Event-averaged measurements of single-molecule kinetics, Chem. Phys. Lett., 2000, 327(1–2), 38–44. 160 X. Chen and R. J. Silbey, Excitation Energy Transfer in a Non-Markovian Dynamical Disordered Environment: Localization, Narrowing, and Transfer Efficiency, J. Phys. Chem. B, 2011, 115(18), 5499–5509. 161 O. Flomenbom, J. Klafter and A. Szabo, What can one learn from two-state single-molecule trajectories?, Biophys. J., 2005, 88(6), 3780–3783. 162 X. Chen and R. J. Silbey, Effect of correlation of local fluctuations on exciton coherence, J. Chem. Phys., 2010, 132(20), 204503. 163 M. K. Prakash and R. A. Marcus, An interpretation of fluctuations in enzyme catalysis rate, spectral diffusion, and radiative component of lifetimes in terms of electric field fluctuations, Proc. Natl. Acad. Sci. U. S. A., 2007, 104(41), 15982–15987. 164 R. H. Zhou, X. H. Huang, C. J. Margulis and B. J. Berne, Hydrophobic collapse in multidomain protein folding, Science, 2004, 305(5690), 1605–1609.


Chem Soc Rev

165 G. G. Hammes, Y.-C. Chang and T. G. Oas, Conformational selection or induced fit: a flux description of reaction mechanism, Proc. Natl. Acad. Sci. U. S. A., 2009, 106(33), 13737–13741. 166 S. M. Sullivan and T. Holyoak, Enzymes with lid-gated active sites must operate by an induced fit mechanism instead of conformational selection, Proc. Natl. Acad. Sci. U. S. A., 2008, 105(37), 13829–13834. 167 T. Wlodarski and B. Zagrovic, Conformational selection and induced fit mechanism underlie specificity in noncovalent interactions with ubiquitin, Proc. Natl. Acad. Sci. U. S. A., 2009, 106(46), 19346–19351. 168 D. Pan, et al., Fluctuating two-state light harvesting in a photosynthetic membrane, J. Phys. Chem. C, 2007, 111(25), 8948–8956. 169 Z. J. Guo, M. Gibson, S. Sitha, S. Chu and U. Mohanty, Role of large thermal fluctuations and magnesium ions in t-RNA selectivity of the ribosome, Proc. Natl. Acad. Sci. U. S. A., 2011, 108(10), 3947–3951. 170 R. Pan, et al., Substrate-induced Changes in Protease Active Site Conformation Impact on Subsequent Reactions with Substrates, J. Biol. Chem., 2010, 285(30), 22948–22954. 171 Q. Zhang, A. C. Stelzer, C. K. Fisher and H. M. Al-Hashimi, Visualizing spatially correlated dynamics that directs RNA conformational transitions, Nature, 2007, 450(7173), 1263–1267. 172 R. D. Astumian and B. Robertson, Imposed Oscillations of Kinetic Barriers Can Cause an Enzyme to Drive a Chemical Reaction Away from Equilibrium, J. Am. Chem. Soc., 1993, 115(24), 11063–11068. 173 J. Blaszczyk, et al., Essential roles of a dynamic loop in the catalysis of 6-hydroxymethyl-7,8-dihydropterin pyrophosphokinase, Biochemistry, 2004, 43(6), 1469–1477. 174 J. Blaszczyk, G. B. Shi, Y. Li, H. G. Yan and X. H. Ji, Reaction trajectory of pyrophosphoryl transfer catalyzed by 6-hydroxymethyl-7,8-dihydropterin pyrophosphokinase, Structure, 2004, 12(3), 467–475.

Chem. Soc. Rev., 2014, 43, 1118--1143 | 1143

Sizing up surfactant synthesis.

Sizing-up the brain.

Single-molecule enzymatic conformational dynamics: spilling out the product molecules.

Sizing-up the brain.

Sizing-up the brain.

Cerebral cortex: Sizing up the columns.

Sizing up the poly(A) tail: insights from deep sequencing.

Protein dynamics: Conformational footprints.

Protein Allostery and Conformational Dynamics.

Generative models of conformational dynamics.

Ketamine dosing in adolescents: sizing up the problem.

Conformational dynamics through an intermediate.

Markov state models of biomolecular conformational dynamics.

Conformational sampling using high-temperature molecular dynamics.

Conformational Dynamics and Allostery in Pyruvate Kinase.

Ligand-Dependent Conformational Dynamics of Dihydrofolate Reductase.

Conformational Dynamics of Ago-Mediated Silencing Processes.

Sizing Up Objects: The Effect of Diminutive Forms on Positive Mood, Value, and Size Judgments.

A Growing Problem: Sizing Up the Burden of Non-melanoma Skin Cancer.

Hierarchical Biomolecular Dynamics: Picosecond Hydrogen Bonding Regulates Microsecond Conformational Transitions.

From residue coevolution to protein conformational ensembles and functional dynamics.

Single-molecule conformational dynamics of a biologically functional hydroxocobalamin riboswitch.

Effect of Redox Partner Binding on Cytochrome P450 Conformational Dynamics.

Allostery and conformational dynamics in cAMP-binding acyltransferases.