Single-molecule imaging inside cells is a valuable tool to study subcellular structures, gene expression, and the dynamics of biomolecules. Here, we present highly inclined swept tile illumination microscopy. By sweeping a thin highly inclined and laminated optical sheet (HILO) with confocal slit detection, our method provides a twofold thinner illumination and greater than fortyfold larger imaging area than conventional HILO microscopy, enabling 3D single-molecule imaging with a high signal-to-background ratio. We demonstrate single-molecule mRNA imaging with a few probes or a single probe in cultured cells and mouse brain tissues, and video-rate live-cell imaging.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Single-molecule imaging is an indispensable tool in many biological studies, i.e., for revealing the dynamics of biomolecules , the ultra-structures of sub-cellular components , and the spatial context of gene expression levels . To visualize individual fluorescent molecules, it is critical to ensure a high signal-to-noise ratio. Although fluorescence signal can be amplified by multiple tagging systems such as SunTag , their large size can potentially interfere with the activity of the target. Therefore, it is more desirable to decrease background level by minimizing unwanted out-of-focus fluorescent signal generated by the excitation beam in microscope systems. Total internal reflection fluorescence (TIRF) microscopy is widely used for single-molecule imaging  due to the fact that it creates an extremely shallow evanescent field, but TIRF or pseudo-TIRF  is only capable of probing the bottom-most of the cell.
Since selective-plane illumination microscopy or light-sheet microscopy naturally provides a good sectioning capability , it has been suggested for 3D single-molecule imaging . However, a configuration of orthogonal excitation and detection objectives limits the usable numerical aperture (NA), resulting in a lower photon collection efficiency that makes it difficult to realize high-contrast single-molecule imaging. To remedy this problem, a micro-mirror [9,10] or a tilted light sheet illuminator  was introduced in the sample chamber to redirect a thin excitation beam; however, this approach complicates the imaging system and demands a special sample preparation. While lattice light-sheet microscopy has demonstrated excellent single-molecule imaging capability , it is made less appealing to many researchers by the complexity of the imaging system, where two water-dipping objectives and Bessel beams are used, and by the fact that it is not compatible with an inverted microscope and conventional sample mounting.
For these reasons, highly inclined and laminated optical sheet (HILO) microscopy [13,14] has been the most common approach for obtaining greater imaging depth in single-molecule imaging, where an incident beam refracts at the glass/water interface with an angle slightly smaller than the critical angle, yielding a thin illumination. This beam passes through the center of the imaging plane and thus allows 3D imaging by reducing out-of-focus background signal. Unlike the aforementioned light-sheet-based approaches [9–12], HILO requires only a single high-numerical-aperture objective without an additional illuminator or reflector, and is compatible with typical sample chambers. These advantages have allowed HILO microscopy to be exploited in several research areas [15,16].
However, in HILO illumination, the beam thickness, , is closely related to the diameter of illumination beam, , i.e., , where is the angle of the transmitted beam [Fig. 1(a)] . This means that a thinner illumination unavoidably results in a smaller imaging area. Therefore, a previous study was only able to demonstrate high contrast imaging with field of view (FOV) and approximately 6–7 μm beam thickness , which makes it challenging to use for imaging a large mammalian cell, multiple cells, or tissues with high contrast. Here, we overcome this limitation of HILO microscopy by sweeping a highly inclined beam elongated in one direction in conjunction with a confocal slit while maintaining the advantages of HILO imaging.
A. High-Contrast Single-Molecule Imaging by Highly Inclined Swept Tile Illumination
First, we created an elongated beam on the conjugated image plane using a pair of cylindrical lenses and sent it to our microscope with a high incidence angle like HILO illumination [Fig. 1(b), Supplement 1, Fig. S1]. For imaging, we prepared a 3D single-molecule hydrogel sample where each DNA probe was labeled with Atto647N and anchored to the hydrogel network via an acrydite moiety during gel polymerization (see Section 4, Methods). We imaged the sample 5 μm above the surface [Figs. 1(c)–1(g)]. While in conventional HILO microscopy an iris controls the size of the illumination beam in both the and directions , our approach generates a tile-like beam elongated along the axis on the sample plane, which is orthogonal to the direction of the illumination beam [Fig. 1(f)]. This tile illumination increased the FOV from to . Importantly, it enabled the visualization of single molecules with higher contrast than HILO illumination with [Fig. 1(e)] because the elongation of the beam in the axis does not affect the beam thickness. In contrast, single-molecule spots could hardly be detected with epi-illumination [Fig. 1(c)] or HILO illumination with a large beam size [Fig. 1(d)] due to the strong background, reconfirming that the image contrast of HILO illumination is highly dependent upon the beam size.
In order to extend the imaging area, we swept the tile along the axis by rotating a galvo mirror [Fig. 1(b)]. However, the highly inclined beam generated out-of-focus background and blurred the image. To resolve this issue, we adopted confocal slit detection using a scientific complementary metal-oxide semiconductor (sCMOS) camera supporting a rolling shutter mode . Synchronously sweeping the tile with the readout of the camera facilitated the rejection of background in real time without additional optical components  (Supplement 1, Fig. S2). We have named our technique highly inclined swept tile (HIST) microscopy.
The size of the total imaging area was decoupled from the beam thickness, which solely depends on the width of tile, and thus it enabled a thinner illumination and larger FOV imaging. Remarkably, we were able to clearly visualize single molecules across FOV [Fig. 1(g)], which was more than 40 times larger than conventional HILO imaging. A line profile clearly showed the much improved signal-to-background ratio (SBR) of HIST microscopy [Fig. 2(a)] and it was times higher than the large-area HILO counterpart [Fig. 2(b)].
B. Characterization of HIST Microscopy
We further characterized the performance of our method using fluorescent beads embedded in 3D hydrogel with 638 nm excitation light. First, we measured the effective illumination width and thickness of the tile at a compression ratio () of the beam that depends on the pair of cylindrical lenses. For the beam thickness measurement, 200 nm beads embedded in hydrogel were used. Upon a tile beam illumination with 638 or 561 nm laser, fluorescence intensities of each bead at different detection planes () were measured by moving a detector with a micrometer. We obtained a corrected depth () considering a longitudinal translation of the sample plane and that of the detector plane, which are related as , where is the refractive index of the sample and is the transverse magnification of the system. The intensity at each depth was averaged from beads, and the full width at half maximum (FWHM) of the intensity profile was used for estimating the beam thickness. For the effective width measurement, we illuminated a tile beam with a compression ratio of 5 or 8 on a 20 nm bead hydrogel sample, and carried out a standard-deviation projection  that effectively excludes out-of-focus background fluorescence. It results in a line profile along the axis, and the plateau area of the profile was used as the beam width. The measured width and thickness were and , respectively, at when a long tile was used [Figs. 3(a)–3(c)]. If a lower compression ratio and/or longer tile is used, the illumination beam becomes proportionally wider and thicker (Supplement 1, Fig. S3). For instance, SBR at was 1.4 times higher than SBR at (Supplement 1, Fig. S4). The beam thickness with 561 nm light was [Fig. 3(d)] and a thin illumination was retained over a depth of [Fig. 3(d)].
Since the tile swept across a large FOV, the illumination angle () was not constant along the axis due to a different refraction angle and aberrations. This resulted in a rather elevated background level on the left side of the HIST images [Fig. 1(g)]. We circumvented this problem in two different ways. First, background subtraction allowed us to readily recover a high-contrast HIST image with a uniform background level (Supplement 1, Fig. S5) . Alternatively, a fine adjustment of the incidence angle by an additional galvo mirror instead of a manual mirror (M) [Fig. 1(b)] helped keep the illumination angle constant during sweeping (Supplement 1, Fig. S6). However, if an intermediate FOV size is used (for example, ), single-galvo-mirror sweeping is adequate for a uniform full FOV imaging.
In addition, we measured the photobleaching kinetics by imaging stacks of imaging volume in order to check whether the eight-times-higher instantaneous illumination intensity of HIST imaging may have adverse effects. The fluorescence intensity of the stack was summed and plotted over time. The decay rate of fluorescence intensity was obtained by a single exponential fit to the time trace, as shown in Supplement 1, Fig. S7. The fluorescence intensity decay rate of HIST imaging was slower than that of epi-illumination at the same average excitation power, presumably because the thin illumination depicted in Fig. 3(c) decreases the amount of accumulated light dose in the imaging volume.
C. Single-Molecule mRNA Imaging in Mammalian Cells with a Single Probe
To demonstrate potential applications of our method, we performed single-molecule RNA fluorescence in situ hybridization (smFISH) with a single probe [20,21] or a few probes , which is critical for detecting single-nucleotide variants  and rare transcriptional mutations. Figure 4(a) displays FISH images of EEF2 (eukaryotic translation elongation factor 2) with four probes labeled with AlexaFluor 647 (AF647) on A549 cells. The HIST image showed higher SBR compared to epi and HILO illumination (Supplement 1, Fig. S8). The photobleaching step distribution revealed the actual number of probes (Supplement 1, Fig. S9). Further, we evaluated the SBR of epi and HIST images with different numbers of FISH probes (, 24, 16, 12, 8, 4, 2, 1) [Figs. 4(b) and 4(c)]. When , the SBR of epi-illumination was not sufficient for robust detection and counting of molecules at our given experimental conditions, i.e., illumination intensity and exposure time 400 ms. Especially for FISH imaging of less than four probes, the autofluorescence background with epi-illumination overwhelmed the fluorescence signal of target mRNAs. In contrast, HIST microscopy showed high SBR () even with a single probe [Figs. 4(c) and 4(d)]. Images with a single probe upon a maximum intensity projection clearly demonstrated the superior performance of HIST microscopy over epi-illumination in Fig. 4(d). In addition, HIST microscopy has a much more uniform illumination profile compared to epi-illumination, which has been shown in scanned light sheet microscopy .
D. Single-Molecule mRNA Imaging in Mouse Brain Tissues with a Few Probes
We imaged EEF2 in 12 μm thick mouse brain tissue with five FISH probes. Compared to adherent cells, tissues are much thicker and have more pronounced autofluorescence background. During our experiment, a 3 μm thick stack of images was acquired at 5.5 μm in depth. It was difficult to distinguish each individual spot with epi-illumination due to high background, while HIST microscopy allowed us to readily detect mRNA [Figs. 5(a) and 5(b)] where SBR was . A control experiment with RNase treatment ensured that our FISH probes bound to the target specifically [Fig. 5(c)].
E. Live-Cell Imaging of Multiple Mammalian Cells
Finally, we imaged actin in living cells using HIST microscopy. We labeled actin in U2OS cells using SiR-actin . While conventional HILO microscopy illuminated a small fraction of the cell, HIST microscopy was able to image a FOV, which allowed us to monitor multiple cells with improved SBR [Figs. 6(a) and 6(b)]. We induced the cell detachment from a coverslip by treatment with trypsin-ethylenediaminetetraacetic acid (EDTA). HIST microscopy recorded at a video rate (23 frames per second) was able to track the shrinkage of U2OS cells in real time [Fig. 6(c)]. An average illumination power was .
Since HILO illumination was first introduced in 2008, it has been used as a primary method, along with TIRF illumination, for single-molecule imaging inside cells and near the cell surface. However, as the beam thickness depends on the size of the illumination area, many imaging techniques taking advantage of HILO illumination have suffered from a limited FOV. This made it particularly impracticable for imaging large tissue samples. Spinning disk confocal microscopy may overcome this problem, but its illumination and detection efficiency is relatively low [4,25]. Despite the advancements of light-sheet microscopy, using a single objective and being readily implemented onto a standard inverted microscope are highly attractive features in single-molecule imaging. Our HIST imaging system provides thinner and much wider illumination compared to HILO imaging. Large FOV with an improved SBR will allow us to study the spatial distribution of biomolecules and their interactions with other cellular components throughout whole cells. Furthermore, it will enable the monitoring of differing cellular responses to stimuli. It also features much higher photon collection efficiency compared to oblique illumination imaging . It will be feasible to obtain much thinner illumination by using a larger compression ratio. We anticipate that HIST microscopy will benefit several applications including super-resolution imaging, single-molecule tracking, and smFISH-based high-throughput gene expression profiling.
A. HIST Microscope
All images were acquired by our custom-made microscope. Three lasers (405, 561, and 638 nm; Cobolt) were coupled to a single-mode fiber (Thorlabs), and their powers were controlled by a combination of a polarizing beam splitter and a half-wave plate. The fiber output was collimated by a lens (L1, ) and sent to a telescope composed of two cylindrical lenses (CL1, ; CL2, ) to generate a tile beam compressed or . The beam was relayed by another telescope system (L2, ; L3, ) with a single-axis galvo mirror (GVS211, Thorlabs), passed through a lens (L4, ), reflected by a dichroic mirror (Di03-R405/488/561/635-t3, Semrock) and focused onto the back focal plane of an objective (PlanApo, , Olympus). The galvo mirror conjugated to the back focal plane was used to sweep the illumination beam on the imaging plane and was controlled by a function generator (DG1032Z, Rigol). A beam incidence angle () was adjusted by a mirror (M) conjugated to the imaging plane. A three-axis piezo stage (MAX311D, Thorlabs) was used for holding samples and acquiring stack images, which was controlled by an analog output board (PCI-6733, National Instruments). Fluorescence emission was collected by the same objective and passed through a filter (FF01-446/523/600/677, Semrock). The emission light was then focused on a sCMOS camera (ORCA-Flash4.0 LT, C11440-22C, Hamamatsu) by a tube lens (). Optionally, a relay system () with a slit at the conjugated imaging plane can be inserted before the tube lens to reduce additional scattered light. The function generator sent a master trigger signal to the camera and piezo stage for synchronizing the image acquisition. The sCMOS camera was run in external rolling shutter trigger mode with a line integration time of 60 ms and a delay time of 0.36 ms per line, which corresponds to 800 ms per frame. See Supplement 1, Fig. S2 for details.
B. Single-Molecule Imaging on 3D Hydrogel
A hydrogel solution was prepared using 7.5% acrylamide: bisacrylamide () (National Diagnostics), 0.2% (v/v) tetramethylethylenediamine (TEMED), and 0.02% (w/v) ammonium persulfate in TAE buffer (tris-acetate-EDTA). An 18 nt single-stranded DNA with an acrydite moiety at and Atto647N at the end (called probe 1) was added to the hydrogel solution at a final concentration of 4 nM, and 50 μL of this mixture was dropped onto a clean coverslip, sandwiched with another coverslip, and incubated for 1 h at room temperature in the dark. After gently separating them, a thin hydrogel layer about 40 μm thick was washed with and incubated in TAE buffer for 1 h to remove unbound DNA. We used an imaging buffer composed of TAE, 0.8% (w/v) dextrose, 1 mg/mL glucose oxidase, 0.04 mg/mL catalase, and 2 mM Trolox during all experiments . All chemicals and DNA were purchased from Sigma-Aldrich and IDT unless specified. Please see Supplement 1, Table 1 for complete oligo sequence information.
C. Imaging Analysis
All images were binned and the pixel size was 130 nm. When obtaining background subtracted images, a local minimum value in a sub-region ( pixels) was calculated and smoothed to obtain a background image which was subtracted from the raw image with an offset to avoid a negative value. For the calculation of the SBR, fluorescence images with different illumination methods were acquired at an imaging depth of . Single-molecule spots were selected using our custom-made scripts using a series of criteria . The fluorescence intensity of each isolated imaging spot was summed in a pixel area around the peak. The corresponding background level (), which was averaged from the surrounding pixels of each selected spot, was used for fluorescence intensity correction. For all the calibrations, SBRs were defined as , where was the sum of the fluorescence intensity of central pixels around the peak divided by the average spot size (from point spread function measurement). In each illumination case, more than 100 independent detectable imaging spots were used for S/B comparison.
D. Fluorescent Nanoparticle Imaging on 3D Hydrogel
Similar to previous experiments, 20 nm diameter crimson beads (ThermoFisher, F8782) were mixed with a 12% hydrogel solution. 50 μL of the mixture was injected into a flow chamber, and after 10 min the image was measured by an industrial CMOS camera (DMK 33UX290, The Imaging Source) with 200 mm focal length tube lens.
E. smFISH on Cultured Mammalian Cells
A549 cells (human lung carcinoma, ATCC CCL-185) were cultured with F-12K medium (ATCC, 30-2004) supplemented with 10% fetal bovine serum (F2442, Sigma) and 1% penicillin/streptomycin (ThermoFisher, 15140122). They were plated on an eight-well Lab-Tek chamber and incubated at 37°C with 5% for 48–72 h. Cells were fixed with 4% (v/v) paraformaldehyde (PFA; Electron Microscopy Sciences, 15710) at room temperature for 10 min. After washing three times with PBS, cells were permeabilized by 0.5% (v/v) Triton X-100 in PBS for 15 min. After again washing three times with PBS, cells were incubated overnight at 37˚C with a various number of FISH probes (, 24, 16, 12, 8, 4, 2, 1) in a hybridization buffer [100 mg/ml dextran sulfate, 1 mg/ml E.coli tRNA (Roche, 10109541001), 2 mM Vanadyl ribonucleoside complex (New England Biolabs, S1402S), 0.2 mg/ml RNase free bovine serum albumin (Ambion, AM2616), SSC, and 10% deionized formamide (Ambion, AM9342)]. The concentration of each probe used was 2.5 nM. The probes against EEF2 were designed by Stellaris Probe Designer, and amine modified probes were purchased from IDT. All the EEF2 probes were labeled with AlexaFluor647 (AF647, Invitrogen, A10277). The nuclei were stained with DAPI. The stack images were obtained at 0.25 μm steps with the illumination intensity of . After maximum projection of the stack images, the SBR was calculated as just described. Additionally, we measured the photobleaching steps of four FISH probes with an illumination power of at a fixed imaging plane. More than 200 time traces at least were used for constructing the distribution of the photobleaching step.
F. smFISH on Mouse Brain Tissues
Wild-type C57BL6 male mice were used in the present study. The mice were transcardially perfused with PBS containing heparin (10 units/mL) and fixed with 4% PFA in 0.1 M phosphate buffer (pH 7.4). Mouse brains were recovered, post-fixed in 4% PFA solution overnight at 4°C, and kept in 30% sucrose in PBS at 4°C until sinking. The brains were coronally sectioned with 12 μm thickness using a cryostat, and collected in PBS. The brain tissue sections were gently transferred to an eight-well Lab-Tek chamber supplemented with 500 μL PBS. The tissue was permeabilized with 200 μL of 0.5% Triton X-100 for 25 min at room temperature. After washing out three times with PBS, the brain tissue was stained with five FISH probes labeled with AF647 overnight in the hybridization buffer at 37˚C. The sample was rinsed with wash buffer (10% deionized formamide in SSC), incubated in SSC for an hour at 37˚C, and supplemented with imaging buffer before imaging. For the nonspecific binding test of FISH probes, we prepared the tissue sample in the same manner except that 0.5% RNase A (TheromoFisher, 12091021) was added to the hybridization buffer. All work with mice was performed in accordance with the Institutional Animal Care and Use Committee guidelines of the University of Central Florida.
G. Live-Cell Imaging on Cultured Mammalian Cells
U2OS cells (human bone osteosarcoma, ATCC HTB-96) were cultured under the same conditions as the A549 cells but using McCoy’s 5a culture medium (Hyclone, AC10250496). After being plated on an eight-well Lab-Tek chamber for more than 36 h, 1 μM SiR-actin (Cytoskeleton, CY-SC001) was incubated for 1 h in the cell culture medium. The chamber was briefly washed before imaging.
University of Central Florida CREOL; Defense Advanced Research Projects Agency (DARPA) (HR00111720066).
We thank Goun Je for preparation of brain tissues and Hunt Optics for generously loaning the sCMOS camera.
See Supplement 1 for supporting content.
1. T. Ha, I. Rasnik, W. Cheng, H. P. Babcock, G. H. Gauss, T. M. Lohman, and S. Chu, “Initiation and re-initiation of DNA unwinding by the Escherichia coli Rep helicase,” Nature 419, 638–641 (2002). [CrossRef]
2. K. Xu, G. Zhong, and X. Zhuang, “Actin, spectrin, and associated proteins form a periodic cytoskeletal structure in axons,” Science 339, 452–456 (2013). [CrossRef]
3. A. Raj, P. van den Bogaard, S. A. Rifkin, A. van Oudenaarden, and S. Tyagi, “Imaging individual mRNA molecules using multiple singly labeled probes,” Nat. Methods 5, 877–879 (2008). [CrossRef]
4. M. E. Tanenbaum, L. A. Gilbert, L. S. Qi, J. S. Weissman, and R. D. Vale, “A protein-tagging system for signal amplification in gene expression and fluorescence imaging,” Cell 159, 635–646 (2014). [CrossRef]
5. T. Funatsu, Y. Harada, M. Tokunaga, K. Saito, and T. Yanagida, “Imaging of single fluorescent molecules and individual ATP turnovers by single myosin molecules in aqueous-solution,” Nature 374, 555–559 (1995). [CrossRef]
6. B. X. Cui, C. B. Wu, L. Chen, A. Ramirez, E. L. Bearer, W. P. Li, W. C. Mobley, and S. Chu, “One at a time, live tracking of NGF axonal transport using quantum dots,” Proc. Natl. Acad. Sci. USA 104, 13666–13671 (2007). [CrossRef]
7. J. Huisken, J. Swoger, F. Del Bene, J. Wittbrodt, and E. H. K. Stelzer, “Optical sectioning deep inside live embryos by selective plane illumination microscopy,” Science 305, 1007–1009 (2004). [CrossRef]
8. F. C. Zanacchi, Z. Lavagnino, M. P. Donnorso, A. Del Bue, L. Furia, M. Faretta, and A. Diaspro, “Live-cell 3D super-resolution imaging in thick biological samples,” Nat. Methods 8, 1047–1049 (2011). [CrossRef]
9. J. C. M. Gebhardt, D. M. Suter, R. Roy, Z. W. Zhao, A. R. Chapman, S. Basu, T. Maniatis, and X. S. Xie, “Single-molecule imaging of transcription factor binding to DNA in live mammalian cells,” Nat. Methods 10, 421–426 (2013). [CrossRef]
10. R. Galland, G. Grenci, A. Aravind, V. Viasnoff, V. Studer, and J.-B. Sibarita, “3D high- and super-resolution imaging using single-objective SPIM,” Nat. Methods 12, 641–644 (2015). [CrossRef]
11. A. K. Gustavsson, P. N. Petrov, M. Y. Lee, Y. Shechtman, and W. E. Moerner, “3D single-molecule super-resolution microscopy with a tilted light sheet,” Nat. Commun. 9, 123 (2018). [CrossRef]
12. B.-C. Chen, W. R. Legant, K. Wang, L. Shao, D. E. Milkie, M. W. Davidson, C. Janetopoulos, X. S. Wu, J. A. Hammer III, Z. Liu, B. P. English, Y. Mimori-Kiyosue, D. P. Romero, A. T. Ritter, J. Lippincott-Schwartz, L. Fritz-Laylin, R. D. Mullins, D. M. Mitchell, J. N. Bembenek, A.-C. Reymann, R. Boehme, S. W. Grill, J. T. Wang, G. Seydoux, U. S. Tulu, D. P. Kiehart, and E. Betzig, “Lattice light-sheet microscopy: imaging molecules to embryos at high spatiotemporal resolution,” Science 346, 1257998 (2014). [CrossRef]
13. M. Tokunaga, N. Imamoto, and K. Sakata-Sogawa, “Highly inclined thin illumination enables clear single-molecule imaging in cells,” Nat. Methods 5, 159–161 (2008). [CrossRef]
14. C. A. Konopka and S. Y. Bednarek, “Variable-angle epifluorescence microscopy: a new way to look at protein dynamics in the plant cell cortex,” Plant J. 53, 186–196 (2008). [CrossRef]
15. A. S. Hansen, I. Pustova, C. Cattoglio, R. Tjian, and X. Darzacq, “CTCF and cohesin regulate chromatin loop stability with distinct dynamics,” eLife 6, e25776 (2017).
16. R. Jungmann, M. S. Avendano, J. B. Woehrstein, M. J. Dai, W. M. Shih, and P. Yin, “Multiplexed 3D cellular super-resolution imaging with DNA-PAINT and Exchange-PAINT,” Nat. Methods 11, 313–318 (2014). [CrossRef]
17. E. Baumgart and U. Kubitscheck, “Scanned light sheet microscopy with confocal slit detection,” Opt. Express 20, 21805–21814 (2012). [CrossRef]
18. R. Jungmann, C. Steinhauer, M. Scheible, A. Kuzyk, P. Tinnefeld, and F. C. Simmel, “Single-molecule kinetics and super-resolution microscopy by fluorescence imaging of transient binding on DNA origami,” Nano Lett. 10, 4756–4761 (2010). [CrossRef]
19. M. A. Reyer, E. L. McLean, S. Chennakesavalu, and J. Y. Fei, “An automated image analysis method for segmenting fluorescent bacteria in three dimensions,” Biochemistry 57, 209–215 (2018). [CrossRef]
20. M. J. Levesque, P. Ginart, Y. C. Wei, and A. Raj, “Visualizing SNVs to quantify allele-specific expression in single cells,” Nat. Methods 10, 865–867 (2013). [CrossRef]
21. G. M. Wadsworth, R. Y. Parikh, J. C. Choy, and H. D. Kim, “mRNA detection in budding yeast with single fluorophores,” Nucleic Acids Res. 45, e141 (2017). [CrossRef]
22. E. Lubeck and L. Cai, “Single-cell systems biology by super-resolution imaging and combinatorial labeling,” Nat. Methods 9, 743–748 (2012). [CrossRef]
23. P. J. Keller, A. D. Schmidt, J. Wittbrodt, and E. H. K. Stelzer, “Reconstruction of zebrafish early embryonic development by scanned light sheet microscopy,” Science 322, 1065–1069 (2008). [CrossRef]
24. G. Lukinavicius, L. Reymond, E. D’Este, A. Masharina, F. Gottfert, H. Ta, A. Guther, M. Fournier, S. Rizzo, H. Waldmann, C. Blaukopf, C. Sommer, D. W. Gerlich, H. D. Arndt, S. W. Hell, and K. Johnsson, “Fluorogenic probes for live-cell imaging of the cytoskeleton,” Nat. Methods 11, 731–733 (2014). [CrossRef]
25. N. A. Hosny, M. Y. Song, J. T. Connelly, S. Ameer-Beg, M. M. Knight, and A. P. Wheeler, “Super-resolution imaging strategies for cell biologists using a spinning disk microscope,” PLoS ONE 8, e74604 (2013). [CrossRef]
26. M. B. Bouchard, V. Voleti, C. S. Mendes, C. Lacefield, W. B. Grueber, R. S. Mann, R. M. Bruno, and E. M. C. Hillman, “Swept confocally-aligned planar excitation (SCAPE) microscopy for high-speed volumetric imaging of behaving organisms,” Nat. Photonics 9, 113–119 (2015). [CrossRef]
27. J. Tang, Y. Sun, S. Pang, and K. Y. Han, “Spatially encoded fast single-molecule fluorescence spectroscopy with full field-of-view,” Sci. Rep. 7, 10945 (2017). [CrossRef]