Multicolor Three-Dimensional Tracking for Single-Molecule

Subscriber access provided by UNIV OF DURHAM

Multicolor 3-dimensional tracking for single-molecule fluorescence resonance energy transfer measurements Aaron M. Keller, Matthew S DeVore, Dominik G. Stich, Dung M. Vu, Timothy Causgrove, and James H. Werner Anal. Chem., Just Accepted Manuscript • Publication Date (Web): 19 Apr 2018 Downloaded from http://pubs.acs.org on April 19, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Multicolor 3-dimensional tracking for single-molecule fluorescence resonance energy transfer measurements Aaron M. Keller,a Matthew S. DeVore,b Dominik G. Stich,c Dung M. Vu,d Timothy Causgrove,e James H. Wernerf,* a

William Jewell College, Department of Chemistry, Liberty, Missouri, 64068. Evangel University, Department of Natural & Applied Sciences, Springfield, Missouri, 65802. c University of Colorado Denver, Anschutz Medical Campus, Aurora, Colorado, 80045. d Los Alamos National Laboratory, Physical Chemistry & Applied Spectroscopy, Los Alamos, New Mexico, 87545. e Texas A&M University Corpus Christi, Department of Physical & Environmental Sciences, Corpus Christi, Texas, 78412 f Los Alamos National Laboratory, Center for Integrated Nanotechnologies, Los Alamos, New Mexico, 87545. KEYWORDS fluorescence resonance energy transfer, single particle tracking, fluorescence microscopy. b

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

Fluorescence Resonance Energy Transfer (FRET) remains a vital and widely-adopted tool in biophysics for characterizing a wide variety of inter- and intra-molecular interactions.1 By labeling the biomolecule(s) of interest with a fluorescent donor and acceptor probe, nanometer-scale dynamics around the Förster radius, the distance corresponding to 50% transfer efficiency, can be readily observed by monitoring the fluorescence emission of the donor and acceptor.2,3,4 Single-molecule FRET (smFRET)5,6,7 offers additional advantages over traditional ensemble methods by characterizing the heterogeneous behavior of subpopulations within the sample8,9,10,11. The single molecule approach also eliminates the need for synchronizing the molecules for kinetic studies, with forward and reverse reaction rates obtained directly from observed smFRET trajectories.12,13,14 To spatially isolate individual biomolecules within the sample of interest for smFRET measurements, several approaches have been developed. One approach involves allowing the labeled sample to freely diffuse at very low (~pM) concentrations through a confocal-based excitation probe volume.8,9 The advantages of this confocal-based approach are that the biomolecule(s) of interest can be studied in a native freelydiffusing environment with high temporal resolution (~ns) afforded by avalanche photodiode point detectors (APDs). A major disadvantage of the confocal-based method is that the

transit time of the biomolecule through the excitation volume is quite short (~ms), thereby allowing only a limited observation, or a so-called “burst” measurement. An alternative approach involves immobilizing the sample at low concentrations on a passivated surface, typically using biotinstreptavidin chemistry, and imaging many spatially-separated molecules using a wide-field electron-multiplying (EM) CCD camera.15 The advantage of this wide-field approach is that it allows extended observation of many separate biomolecules over time (~s), typically only limited by photo-bleaching of the FRET dyes. However, tethering the biomolecule to the surface may generate unwanted interactions with the surface which may significantly perturb the dynamics of interest. Direct tethering of the biomolecule to the surface can be avoided by encapsulating it in a small ~100–200 nm lipid vesicle, but interactions with the liposome may also perturb the dynamics of the biomolecule.14,16,17 In addition, camera-based imaging affords a lower temporal resolution (~ms) compared to the point-based APDs (~ns). Clearly, a smFRET technique that can follow freely diffusing single molecules for extended periods of time with high temporal resolution is desirable. Here, we demonstrate such a method by combining confocal feedback-based 3D single molecule tracking with simultaneous multi-color observations needed for FRET quantification. We note that there are many

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

methods for tracking single fluorescent molecules or particles in three dimensions, which have been recently reviewed by von Diezmann et al.18 Three dimensional single molecule tracking methods can be camera-based or based upon confocal excitation and detection with active feedback to follow the molecule of interest. In camera-based wide-field 3D tracking microscopy methods, out of plane (Z) information can be obtained by the use of multiple image planes, having a point spread function that encodes the Z position, or by very fast 3D whole-cell imaging.19–21 Wide-field based tracking methods have the distinct advantage that a large number of particles can be tracked simultaneously. The temporal resolution in widefield methods is limited by the readout rate of the camera, which is typically tens of ms, but can be pushed to tens of µs with specialized cameras and very bright labels.22 In confocal-based 3D tracking approaches, 3D positional information can be obtained via a spatial modulation of the excitation beam (e.g., in orbital tracking methods) or by having a probe volume with 3D position sensitivity.23–28 One advantage of confocal based tracking approaches is that the temporal resolution of the trajectory is limited by the emission rate of the fluorophore (and not by a camera framing rate) and timeresolved spectroscopy can be performed on the fluorescent target being tracked, with this temporal information often useful in studies of molecular conformation by smFRET.29 Our approach to one-color 3D tracking is based upon a modified confocal microscope design that uses a unique spatial filter design and active closed loop feedback to re-center the particle within four spatially-filtered detection elements. These elements are arranged as a distorted tetrahedron in sample space.25,28,30,31 While tracking the fluorescent particle in 3D, time-correlated single photon counting (TCSPC) can be utilized to measure fluorescence lifetimes and photon antibunching with ~ns temporal resolution.31 We have demonstrated that our instrument is capable of following the 3D motion of a single allergen receptor throughout the entire spatial extent of a mammalian cell (~10 m).27 We have also shown that our system is capable of tracking many types of fluorescent probes, including quantum dots,25,27,31 organic dyes,32 fluorescent proteins,32 and non-blinking giant quantum dots.33 More recently, we have adapted our system to include a spinning disk for simultaneous confocal wide-field fluorescence imaging, and time-gating to suppress background while tracking in live cells.34,35 Our new smFRET 3D tracking instrument expands our system from four to eight detection elements, allowing for two overlapping tetrahedral probe volumes, each detecting separate wavelengths corresponding to the donor (green) and acceptor (red) emission. Below we describe the design and setup of our smFRET 3D tracking microscope and demonstrate its tracking capability through simulations and experiments following freely-diffusing multicolored fluorescent beads in aqueous solution. We also demonstrate our microscope’s ability to track and quantify individual double-stranded DNA molecules undergoing intramolecular smFRET in a viscous glycerol solution.

EXPERIMENTAL SECTION smFRET 3D Tracking Microscope Design The design of the smFRET 3D tracking system is shown schematically in Figure 1 and was based on our one-color 3D tracking system described in detail previously.25,30,36 A pulsed

Page 2 of 15

Figure 1. Layout of excitation and emission paths for the smFRET 3D tracking microscope.

Figure 2. (A) White light images of custom optical fiber pairs used for collection of fluorescence emission. (B) The idealized donor and acceptor probe volumes form a tetrahedron in sample space. The acceptor detection elements are labeled A0 through A3, while the donor detection elements are labeled D0 through D3.


Page 3 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


485 nm diode laser (PicoQuant LDH-P-C-485 laser with a PDL-800-B driver) was directed into the back port of an inverted microscope (Olympus IX-71) and reflected by a long pass dichroic mirror (Semrock, Di01-R488) to a 60x 1.2 NA water-immersion objective (Olympus) for excitation. The back aperture of the objective was intentionally under-filled with the laser beam to slightly broaden the excitation probe volume to approximately twice the lateral dimension of a diffraction-limited spot. The fluorescence emission was collected by the same objective, filtered by the long-pass dichroic and focused using a tube lens (25 cm focal length acromat, Thor Labs). Another long pass dichroic filter (Semrock, FF580-FDi01) was used to separate the FRET donor and acceptor fluorescence emission into two paths. The donor and acceptor fluorescence were further filtered by bandpass filters in each channel (Semrock, FF02-525/40 and FF01-628/40), denoted BP in Figure 1. In each path, labeled either “Donor Detection” or “Acceptor Detection” in Figure 1, the emission was further separated using a 50/50 beam splitter cube (ThorLabs) and focused onto two pairs of custom optical fibers (Polymicro Technologies). As shown in Figure 2A, the diameter of each fiber was 50 µm and each pair had a center-to-center distance of ~55 µm. The transmitted fiber pairs were arranged parallel to the optical table, giving rise to two spatially-filtered detection elements along the x axis, whereas the reflected fiber pairs were arranged perpendicular to the optical table, giving rise to two detection elements along the y axis (Figure 2A). As shown in Figure 1, the distance of each fiber pair was intentionally offset from the focal plane. For example, in the “Donor Detection” path in Figure 1, the transmitted fiber pair has a shorter distance to the beam splitter (FA) compared to the reflected fiber pair (FB), offsetting the two pairs along the optical, or z, axis. In sample space, this corresponds to the detection of a tetrahedral probe volume in both the donor and acceptor emission paths (Figure 2B). We denote the acceptor probe elements as A0, A1, A2, and A3, and the donor probe elements as D0, D1, D2, and D3. Elements with “0” and “1” as subscripts refer to the detectors parallel to the x axis, whereas elements with “2” and “3” subscripts refer to detectors aligned with the y axis. The system is aligned such that the the probe volume of the donor overlaps with that of the acceptor as shown in Figure 2B. For example, D0 should overlap with A0, D1 should overlap with A1, and so on. The experimental measurement of the probe volume size and overlap is described in the Supporting Information. The z-spacing between the x-axis and y-axis detection elements was found to be ~230 nm in sample space (Figure S-1B). The eight optical fibers are coupled to two four-channel single photon counting modules (Excelitas SPCM-AQ4C) for data acquisition. The TTL pulses from these modules were sent to a National Instruments counter-timer board (PCIe6612) to quantify the photon count rates. A second PCIe-6612 board was used to control the photon counting time bin width. The boards were connected with an RTSI cable and programmed using custom LabVIEW RealTime software. To perform 3D single molecule tracking, the sample sits upon a piezo stage (Physique Instrumente P-733.3DD) for fast xyz positioning via proportional-integral-differential (PID) feedback either every 2 ms (for single DNA tracking) or every 5 ms (for fluorescent bead tracking). The piezo stage was tuned to obtain a step response time of less than 2 ms. Using

custom LabVIEW software communicating through a National Instruments PCI-6731 board, the piezo stage moves the sample in 3D space to keep the fluorescent particle centered between all eight detection elements. This occurs when there is an equal count rate on all detectors. To re-center the fluorescent molecule, the PID feedback algorithm applies a step size in x, y, and z, denoted SSx, SSy, and SSz, respectively, as given by Eqns 1, 2, and 3.

=

=

( + ) − ( + ) ( + ) + ( + )

(1)

=

( + ) − ( + ) ( + ) + ( + )

(2)

[( + ) + ( + )] − [( + ) + ( + )] [( + ) + ( + )] + [( + ) + ( + )]

(3)

In relation to Figure 2B, the count rates of the donor elements are denoted D0, D1, D2, and D3, and the count rates of the acceptor elements are denoted A0, A1, A2, and A3. In these equations, kx, ky, and kz represent proportionality constants (the “P” term in PID feedback) in x, y, and z respectively. The PID feedback algorithm is identical to what we have used previously for our one-color 3D tracking system,31 with the exception that the overlapping donor and acceptor elements are summed together. The numerator in these equations specifies the direction (positive or negative) and is also proportional to the magnitude for the step size. For example, in Eqn 1, if (D1 + A1) is greater than (D0 + A0), this means the particle has moved more towards the positive x direction, and therefore a negative SSx must be applied to re-center. Moreover, a greater relative difference between (D1 + A1) and (D0 + A0) would indicate that a greater step size must be applied. In each case, we divide by the sum of the counts on the detectors to account for particles or molecules with different photon count rates.

3D Tracking Fluorescent Beads After alignment of the probe volumes, we tracked 200-nm Tetraspeck Beads (Life Technologies) which have appreciable fluorescence emission in both the donor and acceptor channels due to presence of multiple fluorophores in the beads. About ~4 L of the undiluted bead solution was placed on a coverslip (Fisher, No 1.5) and the sample was excited with a laser power of ~200 nW at a repetition rate of 40 MHz. The 3D tracking was performed in solution ~25 m above the coverslip surface. To make the donor emission count rate comparable the acceptor and to simulate expected intensity ratios for different FRET effiencies, we placed various neutral density (ND) filters before the band pass filter in the donor channel. Preparation of Fluorescent-Labeled DNA and Tracking A double-stranded DNA oligomer capable of intramolecular FRET between a fluorescent donor, Alexa Fluor 488 (AF488), and an acceptor, Alexa Fluor 594 (AF594) was prepared according to a literature procedure.37,38,39 The forward strand, 5'TGTAAAACGAGAGAGCCTAAAACGATC-3', was labeled at the 5’-end with AF594, and the reverse strand, 5'GATCGTTTTAGGCdTCTCTCGTTTTACA-3', was labeld at the deoxyguanosinethymidine (dT) position with AF488. Both labeled and unlabeled DNA strands were purchased through Eurofins MWG Operon. The donor and acceptor fluorophores were attached to the DNA through a six-carbon



Page 4 of 15

The agreement between the particle’s position (red) and the stage position (blue) in the simulation demonstrates that a single particle can be tracked for extended periods (~s) using the donor and acceptor channels for a detected photon count rate of 100 kHz. We estimated the tracking error of the microscope experimentally using immobilized fluorescent beads (Supporting Information). By measuring the fluctuations in stage motion for an immobilized particle (Figure S-2), we found the error in x and y to be on the order of ~20 nm, while the error in z is about ~40 nm. The z axis has significantly higher tracking error compared to x and y which is consistent with our previous one-color 3D tracking system.31,27 It is important to note that this provides a lower limit estimate for the tracking error as the error would increase with fast moving particles and lower photon count rates.

linker and separated by 12 base pairs in the sequence. The DNA strands were annealed by mixing the AF488-labeled reverse strand and the AF594-labeled forward strand at a 1:10 ratio in EB buffer (Qiagen), followed by thermal cycling three times at 95.0, 58.1, 53.1, and 20.0 °C in 3-min intervals at each temperature. Similarly, we also prepared the DNA sample with the donor only by annealing the unlabeled forward strand with the the labeled reverse strand. We denote the doubly-labeled sample capable of smFRET as AF488-DNAAF594 and the singly-labeled sample containing only the donor as AF488-DNA. Figure 3. Random walk simulation of particle (red) in x, y, and z using a diffusion coefficient of 0.1 2/s. The stage position (blue) demonstrates that the particle is successfully tracked in the simulation over a period of 5 s.

For 3D tracking the annealed AF488-DNA-AF594 and AF488-DNA samples were diluted to ~100 pM in a 90% (v/v) glycerol-water solution. About ~10 L of sample was placed onto the coverslip and the 3D tracking was performed in solution ~25 m above the coverslip surface. A laser power of ~20 W at a repition rate of 40 MHz was used to excite the AF488 donor. To improve the tracking duration and increase the number of data points the feedback loop interval was decreased to 2 ms, approximately twice as long as the step response of the piezo stage.

RESULTS AND DISCUSSION 3D Tracking Simulation and Tracking Error We performed simple Monte Carlo simulations (described in the Supporting Information) to see how the tracking microscope could follow a particle undergoing Brownian motion. A simulated random walk with a diffusion coefficient, D, of 0.1 2/s is shown in Figure 3 as a red trajectory in x, y, and z over the course of 5 s. The blue trajectory in Figure 3 represents the stage motion based on our experimental collection efficiencies for all eight detectors (Figure S-1, Supporting Information), the piezo stage step response, and our multicolor tracking algorithm (Eqns 1–3) with kx = 0.35, ky = 0.35, and kz = 0.75. The stage position is updated once every 5 ms.

Tracking of Fluorescent Beads in Solution Freely diffusing fluorescent beads in water were tracked with our multicolor smFRET 3D tracking microscope. Figure 4A–C illustrates the 3D single particle trajectories showing the positions in x, y, and z, as well as the total emission count rate on all detectors. The apparent effiency of energy transfer from the donor to the acceptor, EFRET, was calculated ratiometrically by IA/(IA + ID), where IA is the total acceptor photon counts and ID is the total donor counts. To demonstrate that our microscope can track particles with a range of apparent EFRET values, various ND filters were placed in the donor emission path to attenuate the donor counts. A ND filter of 0.1 placed in front of the donor band pass filter generates an apparent EFRET of ~0.45 as illustrated by a single trajectory (Figure 4D), and by an average per particle histogram of EFRET for 101 trajectories (Figure 4G). As shown by Figure 4E with the ND = 0.3 filter in the donor channel, an apparent EFRET of ~0.55 was observed in both the single particle trajectory (Figure 4E) and the average per particle histogram of 103 trajectories (Figure 4H). This EFRET value of ~0.55 is consistent with that observed for the immobilized bead with the same ND filter (Figure S-2C, Supporting Information) indicating that the EFRET values remain accurate in comparing diffusing vs. immobilized samples. Upon placing a ND 0.6 filter in the donor channel, an apparent EFRET of ~0.69 was observed as shown by the individual trajectory (Figure 4F) and the average per particle EFRET histogram of 50 trajectories (Figure 4I). In short, we have demonstrated that 200 nm fluorescent beads with a range of apparent EFRET values, 0.45–0.69, can be readily tracked in aqueous solution using our multicolor smFRET 3D tracking microscope. The average tracking durations were 6.4 0.7 s for EFRET ~0.45, 5.3 0.5 s for EFRET ~0.55, and 2.4 0.5 s for EFRET ~0.69. The tracking duration seemingly decreases with increasing EFRET but this is due to attenuating the total emission count rate by inserting more opaque ND filters in the donor channel. We note the energy transfer range explored in this manner was limited by the ND filters inserted in the beam path. Whether a particle or molecule can be tracked is determined by the sum of the photon counts on both detectors and not by the ratio of photons in the red or green channels. As demonstrated by Figure 4A, particles with sufficient brightness could be tracked beyond 40 s. In addition to FRET efficiency information, our data also reveal simultaneous insight into the diffusive behavior of particles. From the x, y, z positions in the trajectories, we can calculate the mean squared displacement over time (MSD),


Page 5 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


and compute a diffusion coefficient, D, for each trajectory using the relation for Brownian motion: = /(6∆), where ∆ is the time lag for the displacement. For all EFRET values, we computed an average diffusion coefficient of 1.9 0.1 2/s, which is consistent with that expected by the Stokes-Einstein equation for a 200-nm particle diffusing in water (D=2.14 2/s, as predicted from the Stokes-Einstein relation.) We note that while this single trajectory has a measured diffusion coefficient 3σ away from the predicted value, one generally obtains a distribution of diffusion coefficients in single particle tracking experiments, with the breadth of this distribution broader than the error estimated in determining D from a single MSD analysis.31 We emphasize that in biological environments, a measurement of the diffusion coefficient could be utilized to correlate EFRET values reporting a conformational state or intermolecular interactions with diffusive behavior.40,41

ing a DNA oligomer labeled with a fluorescent donor, AF488, and acceptor, AF594, in a 90% glycerol-water mixture. Figure 5A shows the 3D trajectory of a single AF488-DNA-AF594 molecule tracked for ~1 s. Figure 5B shows the donor (green) and acceptor (red) count rate along with the apparent EFRET (black) over time. In this glycerol-water mixture we measured an average diffusion coefficient for the DNA of 1.24 0.03 2/s. We found that decreasing our microscope feedback loop to 2 ms increased the number of data points and probability of successfully tracking. The average tracking duration, 159 8 ms, was significantly shorter than the fluorescent beads due to lower photon count rate (~30 kHz), and photoinstability of the dyes in the glycerol medium (susceptibility to blinking, photobleaching, etc…). However, this timescale for observing smFRET of a freely diffusing biomolecule is still 100 times longer than that afforded by a conventional confocal burst measurement transit time (~1 ms).

Tracking DNA Undergoing smFRET We demonstrated our 3D tracking microscope’s ability to follow diffusing biomolecules undergoing smFRET by track-

Figure 4. Tracking of 200 nm Tetraspeck Beads in aqueous solution with various ND filters in the donor channel. (A)–(C) 3D trajectories of diffusing beads showing x, y, and z position as well as the total counts per 5 ms. (D)–(F) Total counts per 5 ms of the donor channel (green), acceptor channel (red), and apparent EFRET (black). (G)–(I) Distributions of average EFRET values for (G) 101 trajectories, (H) 103 trajectories, and (I) 50 trajectories. Panels (A), (D), and (G) incorporated a ND = 0.1 filter, panels (B), (E), and (H), incorporated a ND = 0.3 filter, and panels (C), (F), and (I) used a ND = 0.6 filter.



Page 6 of 15

Upon tracking many AF488-DNA-AF594 molecules we were able to compile the average EFRET values for 473 trajectories (Figure 5D). The distribution shows a peak EFRET value of 0.56 0.07 as determined by Gaussian fitting, which is consistent with that observed in the literature for this particular DNA construct (~0.62).37 Upon tracking the DNA construct containing only the donor, AF488-DNA, we see that the average EFRET distribution is significantly lower with a peak value of ~0.25 (Figure 5C). The EFRET distribution is non-zero primarily due to background counts (~7500 kHz count rate summed over all four acceptor channels for each 2 ms period).

Our ability to simultaneously monitor two colors while tracking allows us to observe the molecules for extended periods of time, an advantage of wide-field smFRET measurements, without surface immobilization and with high temporal resolution, advantages afforded by point-based smFRET burst measurements. With the addition of TCSPC to our system, we expect to also simultaneously measure fluorescence lifetimes which could be used to quantify EFRET more reliably than donor and acceptor intensities alone. In addition to measuring smFRET of a freely-diffusing sample for an extended period of time, we also expect to correlate diffusive behavior42 with protein conformational dynamics or interactions in live cells.40,41

CONCLUSIONS

ASSOCIATED CONTENT

In conclusion, we have developed a new 3D tracking microscope capable of measuring FRET of individual particles and biomolecules diffusing in aqueous solution. Our confocal microscope utilized closed-loop feedback with a piezo stage to re-center the particle within two overlapping tetrahedral probe volumes corresponding to the donor and acceptor channels. Our simulations with a diffusing particle showed good agreement between the stage motion and particle position, and our tracking of an immobilized fluorescent bead showed low tracking error, ~20 nm in x and y, and ~40 nm in z. We were able to successfully track freely diffusing 200-nm fluorescent beads in water for several seconds with high temporal resolution (5 ms). We showed that fluorescent particles with a range of EFRET signals (0.45–0.69) could be readily tracked in solution. We were also able to successfully track single DNA molecules undergoing intramolecular FRET in a 90% glycerol-water solution for hundreds of milliseconds at a temporal resolution of 2 ms. This represents a 100-fold improvement in observation time to quantify smFRET at high temporal resolution compared to conventional burst measurements.

Supporting Information Supporting Information includes the experimental measurement of the probe volumes, the tracking error assessment using fluorescent beads, and a section on how the Monte Carlo simulations of the tracking apparatus work. The Supporting Information is available free of charge on the ACS Publications website.

AUTHOR INFORMATION Corresponding Author * e-mail: [email protected]

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT This work was supported through Los Alamos National Laboratory Directed Research and Development (LDRD) and was performed at the Center for Integrated Nanotechnologies, a U.S. Department of Energy, Office of Basic Energy Sciences user facility at Los Alamos National Laboratory (Contract DE-AC5206NA25396).

REFERENCES (1) (2) (3)

(4)

Figure 5. (A) 3D trajectory showing the x, y, and z position as well as the total count rate per 2 ms for AF488-DNAAF594 diffusing in 90% glycerol-water solution. (B) Total counts per 2 ms of the donor channel (green), acceptor channel (red), and apparent EFRET (black) for the trajectory shown in A. (C) Distribution of average EFRET values for 134 AF488-DNA molecules. (D) Distribution of average EFRET values for 473 AF488-DNA-AF594 molecules. The best-fit Gaussian curve is shown for each distribution in red

(5)

(6) (7)

Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.; Springer: New York, NY, 2010. Stryer, L.; Haugland, R. P. Energy Transfer: A Spectroscopic Ruler. Proc. Natl. Acad. Sci. 1967, 58 (2), 719–726. Clegg, R. M. Fluorescence Resonance Energy Transfer and Nucleic Acids. In Methods in Enzymology; Elsevier, 1992; Vol. 211, pp 353–388. Selvin, P. R. Fluorescence Resonance Energy Transfer. In Methods in Enzymology; Elsevier, 1995; Vol. 246, pp 300– 334. Ha, T.; Enderle, T.; Ogletree, D. F.; Chemla, D. S.; Selvin, P. R.; Weiss, S. 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. Roy, R.; Hohng, S.; Ha, T. A Practical Guide to SingleMolecule FRET. Nat. Methods 2008, 5 (6), 507–516. Schuler, B. Single-Molecule FRET of Protein Structure and Dynamics - a Primer. J. Nanobiotechnology 2013, 11 (Suppl 1), S2–S2.


Page 7 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(8)

(9)

(10)

(11)

(12) (13)

(14)

(15) (16)

(17)

(18)

(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26)

Analytical Chemistry Deniz, A. A.; Dahan, M.; Grunwell, J. R.; Ha, T.; Faulhaber, A. E.; Chemla, D. S.; Weiss, S.; Schultz, P. G. Single-Pair Fluorescence Resonance Energy Transfer on Freely Diffusing Molecules: Observation of Förster Distance Dependence and Subpopulations. Proc. Natl. Acad. Sci. U. S. A. 1999, 96 (7), 3670–3675. Schuler, B.; Lipman, E. A.; Eaton, W. A. Probing the FreeEnergy Surface for Protein Folding with Single-Molecule Fluorescence Spectroscopy. Nature 2002, 419 (6908), 743– 747. DeVore, M. S.; Braimah, A.; Benson, D. R.; Johnson, C. K. Single-Molecule FRET States, Conformational Interchange, and Conformational Selection by Dye Labels in Calmodulin. J. Phys. Chem. B 2016, 120 (19), 4357–4364. Keller, A. M.; Benítez, J. J.; Klarin, D.; Zhong, L.; Goldfogel, M.; Yang, F.; Chen, T.-Y.; Chen, P. Dynamic Multibody Protein Interactions Suggest Versatile Pathways for Copper Trafficking. J. Am. Chem. Soc. 2012, 134 (21), 8934–8943. Single-Channel Recording; Sakmann, B., Neher, E., Eds.; Springer US: Boston, MA, 1995. McKinney, S. A.; Joo, C.; Ha, T. Analysis of Single-Molecule FRET Trajectories Using Hidden Markov Modeling. Biophys. J. 2006, 91 (5), 1941–1951. Benítez, J. J.; Keller, A. M.; Chen, P. Nanovesicle Trapping for Studying Weak Protein Interactions by Single-Molecule FRET. In Methods in Enzymology; Elsevier, 2010; Vol. 472, pp 41–60. Zhuang, X. A Single-Molecule Study of RNA Catalysis and Folding. Science 2000, 288 (5473), 2048–2051. Rhoades, E.; Gussakovsky, E.; Haran, G. Watching Proteins Fold One Molecule at a Time. Proc. Natl. Acad. Sci. 2003, 100 (6), 3197–3202. Cisse, I.; Okumus, B.; Joo, C.; Ha, T. Fueling Protein DNA Interactions inside Porous Nanocontainers. Proc. Natl. Acad. Sci. 2007, 104 (31), 12646–12650. von Diezmann, A.; Shechtman, Y.; Moerner, W. E. ThreeDimensional Localization of Single Molecules for SuperResolution Imaging and Single-Particle Tracking. Chem. Rev. 2017, 117 (11), 7244–7275. Ram, S.; Prabhat, P.; Chao, J.; Ward, E. S.; Ober, R. J. High Accuracy 3D Quantum Dot Tracking with Multifocal Plane Microscopy for the Study of Fast Intracellular Dynamics in Live Cells. Biophys. J. 2008, 95 (12), 6025–6043. Thompson, M. A.; Lew, M. D.; Badieirostami, M.; Moerner, W. E. Localizing and Tracking Single Nanoscale Emitters in Three Dimensions with High Spatiotemporal Resolution Using a Double-Helix Point Spread Function. Nano Lett. 2010, 10 (1), 211–218. Chen, B.-C.; Legant, W. R.; Wang, K.; Shao, L.; Milkie, D. E.; Davidson, M. W.; Janetopoulos, C.; Wu, X. S.; Hammer, J. A.; Liu, Z.; et al. Lattice Light-Sheet Microscopy: Imaging Molecules to Embryos at High Spatiotemporal Resolution. Science 2014, 346 (6208), 1257998. Hiramoto-Yamaki, N.; Tanaka, K. A. K.; Suzuki, K. G. N.; Hirosawa, K. M.; Miyahara, M. S. H.; Kalay, Z.; Tanaka, K.; Kasai, R. S.; Kusumi, A.; Fujiwara, T. K. Ultrafast Diffusion of a Fluorescent Cholesterol Analog in Compartmentalized Plasma Membranes. Traffic 2014, 15 (6), 583–612. Levi, V.; Ruan, Q.; Gratton, E. 3-D Particle Tracking in a TwoPhoton Microscope: Application to the Study of Molecular Dynamics in Cells. Biophys. J. 2005, 88 (4), 2919–2928. Cang, H.; Wong, C. M.; Xu, C. S.; Rizvi, A. H.; Yang, H. Confocal Three Dimensional Tracking of a Single Nanoparticle with Concurrent Spectroscopic Readouts. Appl. Phys. Lett. 2006, 88 (22), 223901. Lessard, G. A.; Goodwin, P. M.; Werner, J. H. ThreeDimensional Tracking of Individual Quantum Dots. Appl. Phys. Lett. 2007, 91 (22), 224106. McHale, K.; Berglund, A. J.; Mabuchi, H. Quantum Dot Photon Statistics Measured by Three-Dimensional Particle Tracking. Nano Lett. 2007, 7 (11), 3535–3539.

(27) Wells, N. P.; Lessard, G. A.; Goodwin, P. M.; Phipps, M. E.; Cutler, P. J.; Lidke, D. S.; Wilson, B. S.; Werner, J. H. Time-Resolved Three-Dimensional Molecular Tracking in Live Cells. Nano Lett. 2010, 10 (11), 4732–4737. (28) Keller, A. M.; Ghosh, Y.; DeVore, M. S.; Phipps, M. E.; Stewart, M. H.; Wilson, B. S.; Lidke, D. S.; Hollingsworth, J. A.; Werner, J. H. 3-Dimensional Tracking of Non-Blinking ‘Giant’ Quantum Dots in Live Cells. Adv. Funct. Mater. 2014, 24 (30), 4796–4803. (29) Werner, J. H.; Joggerst, R.; Dyer, R. B.; Goodwin, P. M. A Two-Dimensional View of the Folding Energy Landscape of Cytochrome C. Proc. Natl. Acad. Sci. 2006, 103 (30), 11130–11135. (30) Lessard, G. A.; Goodwin, P. M.; Werner, J. H. ThreeDimensional Tracking of Fluorescent Particles. In Proc. SPIE, Ultrasensitive and Single-Molecule Detection Technologies; Enderlein, J., Gryczynski, Z. K., Eds.; San Jose, CA, 2006; Vol. 6092, p 609205. (31) Wells, N. P.; Lessard, G. A.; Werner, J. H. Confocal, ThreeDimensional Tracking of Individual Quantum Dots in HighBackground Environments. Anal. Chem. 2008, 80 (24), 9830–9834. (32) Han, J. J.; Kiss, C.; Bradbury, A. R. M.; Werner, J. H. TimeResolved, Confocal Single-Molecule Tracking of Individual Organic Dyes and Fluorescent Proteins in Three Dimensions. ACS Nano 2012, 6 (10), 8922–8932. (33) Keller, A. M.; Ghosh, Y.; DeVore, M. S.; Phipps, M. E.; Stewart, M. H.; Wilson, B. S.; Lidke, D. S.; Hollingsworth, J. A.; Werner, J. H. 3-Dimensional Tracking of Non-Blinking ‘Giant’ Quantum Dots in Live Cells. Adv. Funct. Mater. 2014, 24 (30), 4796–4803. (34) DeVore, M. S.; Stich, D. G.; Keller, A. M.; Ghosh, Y.; Goodwin, P. M.; Phipps, M. E.; Stewart, M. H.; Cleyrat, C.; Wilson, B. S.; Lidke, D. S.; et al. Three Dimensional TimeGated Tracking of Non-Blinking Quantum Dots in Live Cells. In Proc. SPIE, Colloidal Nanoparticles for Biomedical Applications X; Parak, W. J., Osinski, M., Liang, X.-J., Eds.; San Francisco, CA, 2015; Vol. 9338, p 933812. (35) DeVore, M. S.; Stich, D. G.; Keller, A. M.; Cleyrat, C.; Phipps, M. E.; Hollingsworth, J. A.; Lidke, D. S.; Wilson, B. S.; Goodwin, P. M.; Werner, J. H. Note: Time-Gated 3D Single Quantum Dot Tracking with Simultaneous Spinning Disk Imaging. Rev. Sci. Instrum. 2015, 86 (12), 126102. (36) Wells, N. P.; Lessard, G. A.; Goodwin, P. M.; Phipps, M. E.; Cutler, P. J.; Lidke, D. S.; Wilson, B. S.; Werner, J. H. Time-Resolved Three-Dimensional Molecular Tracking in Live Cells. Nano Lett. 2010, 10 (11), 4732–4737. (37) DeVore, M. S.; Gull, S. F.; Johnson, C. K. Classic Maximum Entropy Recovery of the Average Joint Distribution of Apparent FRET Efficiency and Fluorescence Photons for Single-Molecule Burst Measurements. J. Phys. Chem. B 2012, 116 (13), 4006–4015. (38) Widengren, J.; Schweinberger, E.; Berger, S.; Seidel, C. A. M. Two New Concepts to Measure Fluorescence Resonance Energy Transfer via Fluorescence Correlation Spectroscopy: Theory and Experimental Realizations. J. Phys. Chem. A 2001, 105 (28), 6851–6866. (39) Antonik, M.; Felekyan, S.; Gaiduk, A.; Seidel, C. A. M. Separating Structural Heterogeneities from Stochastic Variations in Fluorescence Resonance Energy Transfer Distributions via Photon Distribution Analysis. J. Phys. Chem. B 2006, 110 (13), 6970–6978. (40) Ma, Y.; Pandzic, E.; Nicovich, P. R.; Yamamoto, Y.; Kwiatek, J.; Pageon, S. V.; Benda, A.; Rossy, J.; Gaus, K. An Intermolecular FRET Sensor Detects the Dynamics of T Cell Receptor Clustering. Nat. Commun. 2017, 8, 15100. (41) Coban, O.; Zanetti-Dominguez, L. C.; Matthews, D. R.; Rolfe, D. J.; Weitsman, G.; Barber, P. R.; Barbeau, J.; Devauges, V.; Kampmeier, F.; Winn, M.; et al. Effect of Phosphorylation on EGFR Dimer Stability Probed by Single-Molecule



Dynamics and FRET/FLIM. Biophys. J. 2015, 108 (5), 1013–1026. (42) Saxton, M. J.; Jacobson, K. Single-Particle Tracking: Applications to Membrane Dynamics. Annu. Rev. Biophys. Biomol. Struct. 1997, 26 (1), 373–399.


Page 8 of 15

Page 9 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC Figure




Page 10 of 15

Page 11 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60





Page 12 of 15

Page 13 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60





Page 14 of 15

Page 15 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60



Multicolor Three-Dimensional Tracking for Single-Molecule

Recommend Documents