Multicolor TIRF LOCALIZATION Microscopy
In order to study biological molecules, we turn to Total Internal Reflection Fluorescence (TIRF) microscopy. By limiting our illumination volume to the surface (~100nm) of a coverslip, we can increase our signal-to-noise ratio such that we can detect single molecules. Our TIRF microscope set ups are capable of four-color, time-shared imaging, which enables us to track multiple components of the cell's microtubule-based transport machinery at the same time.
In conventional light microscopy, the image of a point-like object is a diffraction-limited spot, which has a width of half the wavelength of the emitted light (~250 nm for visible emission). The center of the spot, which represents the position of the probe, can in principle be localized with an arbitrarily high precision by collecting sufficient number of photons from a fluorophore. By collecting more than 10,000 photons per frame, single molecules can be tracked at 1 nm precision. Time resolution of this method is typically 30 ms for organic dyes (Cy3, TMR, Cy5) and a few ms for quantum dot probes. The technique is named Fluorescence Imaging with One Nanometer Accuracy (FIONA).
Using this method, we track the motility of single motors at high precision. These experiments reveal the size and direction of individual steps taken by the motor. By labeling the two heads with a different color, we also determine how the heads are positioned and move relative to each other during processive runs.
A stepping trace of yeast cytoplasmic dynein labeled with QD-585 and QD-655 shows that the heads step independently of each other, a mechanism fundamentally distinct from coordinated hand-over-hand stepping of kinesin and myosin motors (figure from Dewitt et al., Science 2012).
FIONA. (Left) Diffraction-limited image of a single dye on a CCD camera. (Right) The width of the spot is roughly equivalent to half the wavelength of the emitted light. The spot was fitted to a 2-D Gaussian to locate its center with 1.5 nm precision.
Optical Trapping
Optical trapping uses a tightly focused laser beam to manipulate microscopic particles and measure forces and motions at the single-molecule level. Our optical trap uses a 1064 nm infrared laser to apply pN-scale forces while tracking molecular positions with sub-nm precision at sampling rates exceeding 20 kHz. For perspective, 1 pN is approximately the gravitational attraction between two average-sized people standing half a mile apart.
Built around a modified Nikon Ti-E microscope, our instrument incorporates a custom z-focus feedback system to minimize axial drift. It is housed in an acoustically isolated, temperature-controlled room to suppress environmental vibrations that could interfere with sensitive measurements. The trap is fully automated and remotely operated using custom LabVIEW software. Acousto-optic deflectors control laser positioning with updates every 50 microseconds, while an FPGA enables high-speed force feedback and accurate detector calibration.
We use optical trapping to investigate the stepping dynamics and mechanochemistry of molecular motors, particularly dynein, under controlled mechanical loads. The system also integrates two TIRF excitation lines (488 and 633 nm) and a sensitive Andor EMCCD camera, allowing single-molecule fluorescence imaging alongside force measurements.
Reflection of a well-aligned trapping laser off the bottom surface of the coverslip.
Differential detection stage. Splitting the trapping laser beam into orthogonal polarizations allows us to create two traps simultaneously and detect the positions of two trapped beads independently from one another. Differential measurement experiments allow us to de-couple the observed signal from noise sources common to both trapping beams, such as vibrations of the microscope body or stage drift.
Single-Molecule FRET
Single-molecule fluorescence resonance energy transfer (smFRET) is a powerful technique that enables us to observe interactions between proteins and nucleic acids, the structure of these complexes as well as their dynamic conformational changes at single molecule level. We label the samples site-specifically with a pair of fluorophores (called donor-acceptor pair) such that the emission spectrum of the donor overlaps with the excitation spectrum of the donor. Because of this overlap, when the donor is brought to its excited state with a laser, the excitation can be transferred to the acceptor, resulting in the emission of an acceptor photon. Since FRET efficiency is a function of the donor-acceptor separation, we can measure distances in the 2-10 nm range on a time scale ranging from milliseconds to several minutes.
FRET trajectories of Cas-9 binding to a matching DNA target and a DNA target with 1-3bp mismatch. Vertical lines represent acceptor photobleaching.
MINFLUX
MINFLUX addresses the outstanding challenges in spatiotemporal resolution in fluorescence microscopy. In contrast to camera-based single-molecule localization microscopy, MINFLUX is a beam-scanning technique in which the fluorophore positions are determined using predefined scanning patterns encircling each fluorophore. Molecules are excited using a focus shape featuring an intensity zero in the center. From the predefined positions of that zero intensity, the known shape of the excitation focus, and the detected number of photons emitted by the molecule at the different positions of the excitation beam, the position of the molecule can be determined precisely. Using the MINFLUX approach, 2 nm localization precision can be achieved using fewer photons and lower light doses compared to classical tracking techniques. This method closes the prevalent resolution gap from ~20 to 30 nm in STED, PALM/STORM, and other fluorescence nanoscopies to the 1–5 nm size scale of the molecules themselves.
Localization accuracy of MINFLUX. The position of the dye is determined by its relative position to the excitation minimum of a donut. Localization accuracy (σ) depends on the diameter of the donut's zero-intensity "hole" (L), which is typically 70 nm, much smaller than the emission wavelength of the excitation beam (600 nm). Therefore, this method requires 100x fewer photons for nm localization.
CRYO-EM
We use cryo-electron microscopy (cryo-EM) to determine high-resolution structures of microtubules, microtubule-associated proteins (MAPs), and molecular motors. By imaging biological samples rapidly frozen in a thin layer of vitreous ice, cryo-EM enables us to visualize macromolecular structures and interactions in near-native conditions, often at near-atomic resolution. This powerful technique allows us to investigate how motor proteins such as kinesin and dynein interact with microtubules, how MAPs recognize and regulate the microtubule lattice, and how structural changes influence motor activity and cytoskeletal organization.
Helical reconstruction of microtubules decorated with dynein's microtubule-binding domain.