Microscopy, 2015, 227–236 doi: 10.1093/jmicro/dfv036 Advance Access Publication Date: 6 July 2015
Review
STED microscopy—super-resolution bio-imaging utilizing a stimulated emission depletion Kohei Otomo1, Terumasa Hibi1,2, Yuichi Kozawa3, and Tomomi Nemoto1,2,* 1
*To whom correspondence should be addressed. E-mail: [email protected] Received 6 April 2015; Accepted 8 June 2015
Abstract One of the most popular super-resolution microscopies that breaks the diffraction barrier is stimulated emission depletion (STED) microscopy. As the optical set-up of STED microscopy is based on a laser scanning microscopy (LSM) system, it potentially has several merits of LSM like confocal or two-photon excitation LSM. In this article, we first describe the principles of STED microscopy and then describe the features of our newly developed two-photon excitation STED microscopy. On the basis of our recent results and those of other researchers, we conclude by discussing future research and new technologies in this field. Key words: optical vortex, liquid crystal devices, two-photon microscopy
Introduction Fluorescence microscopy is widely used in medical or biological research [1]. Several techniques for visualizing biological events have been developed to meet biological researchers’ expectations. Among these techniques, superresolution microscopy that breaks the spatial resolution limitations of fluorescence microscopy has been a promising recent development [2]. There are several definitions of the spatial resolution in optical microscopy; one of the most famous was presented by the German physicist Ernst Abbe [3]. Abbe developed the diffraction-limited resolution theory and described the lateral spatial resolution as the ratio of the wavelength of light to the numerical aperture of a focusing lens. According to this law, the limit of lateral spatial resolution for optical microscopy using a particular wavelength
of light is approximately one-half of that wavelength. As most of the fluorophores used for bio-imaging have spectral properties in the visible region, it was believed for a long time that visualizations of nanostructures smaller than ∼200 nm were impossible. Therefore, other methods, such as electron microscopy and atomic force microscopy, were used to capture images of biological nanostructures. However, these methods cannot visualize the internal structures of living specimens as fluorescence microscopy can. The biological field thus faced an enormous dilemma in that any existing microscopies could not be visualized the interior of nanostructures of living specimens. However, in the past decade, super-resolution microscopes, that overcame this dilemma by effectively utilizing several principles of optics, physics or chemistry, have been made commercially available by many
© The Author 2015. Published by Oxford University Press on behalf of The Japanese Society of Microscopy. All rights reserved. For permissions, please e-mail: [email protected]
227
Downloaded from http://jmicro.oxfordjournals.org/ by guest on November 16, 2015
Research Institute for Electronic Science, Hokkaido University, Kita 20 Nishi 10, Kita, Sapporo 001-0020, Japan, 2Graduate School of Information Science and Technology, Hokkaido University, Kita 14 Nishi 9, Kita, Sapporo 060-0814, Japan, and 3Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Katahira 2-1-1, Aoba-ku, Sendai 980-8577, Japan
228
companies and implemented in biological research. In recent years, super-resolution microscopies have not only been merely demonstrated technologically but have also yielded new scientific findings. The fact that Dr Eric Betzig of the Howard Huges Medical Institute, Dr William E. Moerner of Stanford University and Stefan W. Hell of the Max Planck Institute for Biophysical Chemistry were awarded the 2014 Nobel Prize in Chemistry is reflective of the great expectations that biological researchers have for this technology.
Stimulated emission depletion microscopy
Fig. 1. (a) Schematic of a conventional STED microscopy system. (b) Focal spots of the excitation light and the STED light, and the acquired STED image.
separate the stimulated emission light, the fluorescent light from the central spot of the doughnut-shaped focal pattern can be selected. The spatial resolution of STED microscopy is expressed approximately as [0.5λ/NA (1 + Isted/Isat)0.5], where λ is the wavelength of light, NA is the numerical aperture of the focusing lens, Isted is the laser power of a STED light and Isat is the standard laser power reflecting the saturation degree of a stimulated emission process [4]. Isat depends on the excitation laser power, the excitation efficiency and the quantum yield of a targeted fluorophore, and the stimulated emission efficiency determined by the relationship between the wavelength of a STED light and a fluorophore. Since the number of molecules in an electronic excited state is finite, the number of molecules returning to an electronic ground state by stimulated emission processes reaches a plateau as the power of a STED light increases. Then, due to the saturation of stimulated emission, the effective size of the central hollow region of a doughnutshaped STED light becomes smaller as the STED power increases (Fig. 2), resulting the diffraction unlimited spatial resolution of STED microscopy. Since the lateral spatial resolution of STED microscopy depends on the size of its central hollow region, there is no limit to the resolution in principle. By using a nitrogen-vacancy centre in diamond as a fluorophore, visualization of nanostructure 0) were focused with a steep dark spot. On the other hand, the central spots of the foci with topological charge smaller than zero (m < 0) tended to be buried. Our results were consistent with previously reported numerical calculations [7]. In this study, we used the optical vortex with a topological charge of one as the STED light, because it had the narrowest dark spot. The focal planes of the two-photon excitation light and STED light used in our system were verified. Figure 7a shows orthogonal views of the xyz-fluorescent images of two different types of fluorescent beads that were placed on the same cover slip. Each of their focal planes was made to become nearly identical by the simple introduction of these laser beams in parallel into the corresponding laser inputs of the microscopy system. Next, to evaluate the spatial resolution of our TP-STED microscopy system, fluorescent images of green beads with diameters of 0.10 μm were acquired with both a conventional TPLSM system and our TP-STED microscope. The two-photon excitation laser power at the output of the microscope objective was 1.3 mW for both TPLSM and TP-STED. The STED laser power at the same position was 89 mW. The pixel size of both images was 2.8 × 2.8 nm, and the pixel dwell time was 12.5 μs. Two representative TPLSM
Fig. 6. Focal patterns of various circularly polarized optical vortices created by tLCDs. Upper panels indicate fluorescence images of a fluorescent bead directly excited by various optical vortices. Lower panels indicate theoretical phase distributions of the optical vortices generated by tLCD-24.
Downloaded from http://jmicro.oxfordjournals.org/ by guest on November 16, 2015
shape at the position of a specimen was compensated by a negative chirper using a pair of prisms. The light source for the STED process was an optically pumped, 577 nm, semiconductor continuous-wave (CW) laser. Spatial mode cleaning of the STED light was accomplished using a spatial filter. The linear polarization direction for STED light was oriented towards the LC molecules in the tLCDs by passing through a half-wave plate. A two-photon excitation light and a STED light, which were modulated by tLCDs as noted below, were merged at the first dichroic mirror and introduced into a Galvano mirror scanner and an upright microscope. Both beams were reflected by the second dichroic mirror and were focused on a specimen using a water immersion objective lens (NA 1.2) of which correction collar was adjusted by evaluating PSFs of optical vortices. Fluorescent light was collected by the objective lens, passed through the second dichroic mirror and emission filters and finally detected with a photomultiplier tube. We used two types of tLCDs; one was a plain cell tLCD (tLCD-P) with homogeneously aligned LC molecules functioning as an applied voltage-dependent variable wave plate. The other was a 24-piece divided phase mask (tLCD-24) that enabled production of a spiral-phase distribution with 24 steps around the centre of the beam, as shown in the lower row of Fig. 6. For STED microscopy, the focus of an optical vortex should have a steep central dark spot, leading to a higher spatial resolution as the size of the centre of the dark spot is reduced [2]. As described in a previous section, to focus an optical vortex, the handedness of the circular polarization was chosen so as to create a fine dark spot at the focus [7]. However, for the case of a circularly polarized optical vortex at the position of the tLCDs, a difference in phase shifts between the s- and p-polarized light generated in the
231
232
Microscopy, 2015, Vol. 64, No. 4
and TP-STED images in Fig. 7b were averages of four and eight acquired images, respectively. After applying the STED light, the fluorescent bead image that had exhibited an isotropic shape in the focal plane became elliptical as the size itself became smaller. We estimated that the FWHM values were 322 and 173 nm for the averaged fluorescence intensity profiles acquired with conventional TPLSM and our TP-STED, respectively. To estimate the PSFs and the spatial resolution of the microscopy more accurately, measurements of the fluorescent images of smaller-sized beads would be required. However, the sensitivity of our conventional TPLSM system was insufficient to clearly visualize such small fluorescent beads, due to weak fluorescence. Finally, we demonstrated that our method was applicable to biological specimens by observing microtubules in fixed COS-7 cells that were immunostained with fluorescent dye-marked antibodies. ATTO 425 was chosen as the fluorescent dye, because the wavelength of our STED light hardly overlapped with its absorption band. The excitation laser power at the output of the microscope objective was 3.4 mW for both imaging. The STED laser power at the same position was 72 mW. The pixel size of both images was 41.4 × 41.4 nm, and the pixel dwell time was 10 μs. As shown in Fig. 8, TP-STED microscopy enabled us to visualize the fine network structures of microtubules more clearly than conventional TPLSM.
Summary and outlook As described in the previous section, by adding tLCDs and a CW laser to our conventional TPLSM system, we developed
a new TP-STED microscopy system with the following useful properties. First, by adjusting the voltage applied to a tLCD-P, we compensated for various phase shifts in the optical path to acquire a fine doughnut-shaped PSF and confirmed that this optical vortex, which had a 24-part helical phase distribution of 2π created by a tLCD-24, was sufficient for STED microscopy. Second, by modifying a conventional TPLSM system by the addition of tLCDs and a depletion laser light source, we developed a TP-STED microscopy system. In principle, this methodology is applicable not only for TPLSM but also for other types of laser scanning microscopies. Therefore, tLCD is expected to achieve super-resolution microscopy more readily. Third, our TP-STED microscopy system was employable at different wavelengths of STED laser light by only tuning the voltages applied to tLCDs. We confirmed that the same tLCDs enabled both 473 and 577 nm laser beams to be converted into optical vortices [9]. One feature of TPLSM is that various fluorescent dyes can be excited simultaneously using a single wavelength NIR light [11,21]. Thus, combined with a wavelength-tunable light source for STED processes, this methodology might allow for super-resolution microscopy to visualize multiple components in living specimens using a variety of fluorophores. Finally, the other type of tLCDs enables to convert a Gaussian beam into a vector beam, which has an inhomogeneous polarization distribution on the beam cross-sections [22,23]. For example, a radially polarized beam, classified as one of the vector beams, is known to generate a strong axial component of the electric field at the focus and has been applied to the molecular
Downloaded from http://jmicro.oxfordjournals.org/ by guest on November 16, 2015
Fig. 7. (a) Merged fluorescence image of green fluorescent beads with diameters of 0.10 μm (green) and orange fluorescent beads with diameters of 0.17 μm (magenta) placed on the same cover slip. (b) Comparisons of the TPLSM and TP-STED images of a green fluorescent bead with a diameter of 0.10 μm. The lower panels show the averaged fluorescence intensity profiles of 10 beads across the intensity centre along the red dashed lines in the fluorescent images. The inlet length value indicates the full width at half maximum.
Microscopy, 2015, Vol. 64, No. 4
233
orientation imaging [24]. In most STED microscopies, a Gaussian beam and an optical vortex with a topological charge of one are used as an excitation light and a STED light, respectively. These two beams have circular polarization at the focus, so that acquired STED images have no information about molecular orientations. On the other hand, a super-resolution molecular orientation imaging using STED methodology has been demonstrated using an azimuthally polarized STED light although only the lateral orientation was determined [25]. If a radially polarized excitation beam and a STED light possessing strong axial electric field at the focus are also used, super-resolution microscopy acquiring three-dimensional molecular orientations can be achieved. This methodology is expected to visualize the detailed nanostructure of the micro-domain in the plasma membrane and may be feasible by our tLCD technologies. The spatial resolution of our present TP-STED microscopy system was, however, estimated to be larger than 100 nm, although the spatial resolution of the TP-STED systems in previous studies were reported to be smaller than 100 nm [16–19]. The reason might be that we used a CW laser as the STED light source; the STED efficiency was reported to be superior under pulse-type STED microscopy [18,19]. However, Willig et al. [26] demonstrated CW STED microscopy visualized nanostructures
Review
STED microscopy—super-resolution bio-imaging utilizing a stimulated emission depletion Kohei Otomo1, Terumasa Hibi1,2, Yuichi Kozawa3, and Tomomi Nemoto1,2,* 1
*To whom correspondence should be addressed. E-mail: [email protected] Received 6 April 2015; Accepted 8 June 2015
Abstract One of the most popular super-resolution microscopies that breaks the diffraction barrier is stimulated emission depletion (STED) microscopy. As the optical set-up of STED microscopy is based on a laser scanning microscopy (LSM) system, it potentially has several merits of LSM like confocal or two-photon excitation LSM. In this article, we first describe the principles of STED microscopy and then describe the features of our newly developed two-photon excitation STED microscopy. On the basis of our recent results and those of other researchers, we conclude by discussing future research and new technologies in this field. Key words: optical vortex, liquid crystal devices, two-photon microscopy
Introduction Fluorescence microscopy is widely used in medical or biological research [1]. Several techniques for visualizing biological events have been developed to meet biological researchers’ expectations. Among these techniques, superresolution microscopy that breaks the spatial resolution limitations of fluorescence microscopy has been a promising recent development [2]. There are several definitions of the spatial resolution in optical microscopy; one of the most famous was presented by the German physicist Ernst Abbe [3]. Abbe developed the diffraction-limited resolution theory and described the lateral spatial resolution as the ratio of the wavelength of light to the numerical aperture of a focusing lens. According to this law, the limit of lateral spatial resolution for optical microscopy using a particular wavelength
of light is approximately one-half of that wavelength. As most of the fluorophores used for bio-imaging have spectral properties in the visible region, it was believed for a long time that visualizations of nanostructures smaller than ∼200 nm were impossible. Therefore, other methods, such as electron microscopy and atomic force microscopy, were used to capture images of biological nanostructures. However, these methods cannot visualize the internal structures of living specimens as fluorescence microscopy can. The biological field thus faced an enormous dilemma in that any existing microscopies could not be visualized the interior of nanostructures of living specimens. However, in the past decade, super-resolution microscopes, that overcame this dilemma by effectively utilizing several principles of optics, physics or chemistry, have been made commercially available by many
© The Author 2015. Published by Oxford University Press on behalf of The Japanese Society of Microscopy. All rights reserved. For permissions, please e-mail: [email protected]
227
Downloaded from http://jmicro.oxfordjournals.org/ by guest on November 16, 2015
Research Institute for Electronic Science, Hokkaido University, Kita 20 Nishi 10, Kita, Sapporo 001-0020, Japan, 2Graduate School of Information Science and Technology, Hokkaido University, Kita 14 Nishi 9, Kita, Sapporo 060-0814, Japan, and 3Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Katahira 2-1-1, Aoba-ku, Sendai 980-8577, Japan
228
companies and implemented in biological research. In recent years, super-resolution microscopies have not only been merely demonstrated technologically but have also yielded new scientific findings. The fact that Dr Eric Betzig of the Howard Huges Medical Institute, Dr William E. Moerner of Stanford University and Stefan W. Hell of the Max Planck Institute for Biophysical Chemistry were awarded the 2014 Nobel Prize in Chemistry is reflective of the great expectations that biological researchers have for this technology.
Stimulated emission depletion microscopy
Fig. 1. (a) Schematic of a conventional STED microscopy system. (b) Focal spots of the excitation light and the STED light, and the acquired STED image.
separate the stimulated emission light, the fluorescent light from the central spot of the doughnut-shaped focal pattern can be selected. The spatial resolution of STED microscopy is expressed approximately as [0.5λ/NA (1 + Isted/Isat)0.5], where λ is the wavelength of light, NA is the numerical aperture of the focusing lens, Isted is the laser power of a STED light and Isat is the standard laser power reflecting the saturation degree of a stimulated emission process [4]. Isat depends on the excitation laser power, the excitation efficiency and the quantum yield of a targeted fluorophore, and the stimulated emission efficiency determined by the relationship between the wavelength of a STED light and a fluorophore. Since the number of molecules in an electronic excited state is finite, the number of molecules returning to an electronic ground state by stimulated emission processes reaches a plateau as the power of a STED light increases. Then, due to the saturation of stimulated emission, the effective size of the central hollow region of a doughnutshaped STED light becomes smaller as the STED power increases (Fig. 2), resulting the diffraction unlimited spatial resolution of STED microscopy. Since the lateral spatial resolution of STED microscopy depends on the size of its central hollow region, there is no limit to the resolution in principle. By using a nitrogen-vacancy centre in diamond as a fluorophore, visualization of nanostructure 0) were focused with a steep dark spot. On the other hand, the central spots of the foci with topological charge smaller than zero (m < 0) tended to be buried. Our results were consistent with previously reported numerical calculations [7]. In this study, we used the optical vortex with a topological charge of one as the STED light, because it had the narrowest dark spot. The focal planes of the two-photon excitation light and STED light used in our system were verified. Figure 7a shows orthogonal views of the xyz-fluorescent images of two different types of fluorescent beads that were placed on the same cover slip. Each of their focal planes was made to become nearly identical by the simple introduction of these laser beams in parallel into the corresponding laser inputs of the microscopy system. Next, to evaluate the spatial resolution of our TP-STED microscopy system, fluorescent images of green beads with diameters of 0.10 μm were acquired with both a conventional TPLSM system and our TP-STED microscope. The two-photon excitation laser power at the output of the microscope objective was 1.3 mW for both TPLSM and TP-STED. The STED laser power at the same position was 89 mW. The pixel size of both images was 2.8 × 2.8 nm, and the pixel dwell time was 12.5 μs. Two representative TPLSM
Fig. 6. Focal patterns of various circularly polarized optical vortices created by tLCDs. Upper panels indicate fluorescence images of a fluorescent bead directly excited by various optical vortices. Lower panels indicate theoretical phase distributions of the optical vortices generated by tLCD-24.
Downloaded from http://jmicro.oxfordjournals.org/ by guest on November 16, 2015
shape at the position of a specimen was compensated by a negative chirper using a pair of prisms. The light source for the STED process was an optically pumped, 577 nm, semiconductor continuous-wave (CW) laser. Spatial mode cleaning of the STED light was accomplished using a spatial filter. The linear polarization direction for STED light was oriented towards the LC molecules in the tLCDs by passing through a half-wave plate. A two-photon excitation light and a STED light, which were modulated by tLCDs as noted below, were merged at the first dichroic mirror and introduced into a Galvano mirror scanner and an upright microscope. Both beams were reflected by the second dichroic mirror and were focused on a specimen using a water immersion objective lens (NA 1.2) of which correction collar was adjusted by evaluating PSFs of optical vortices. Fluorescent light was collected by the objective lens, passed through the second dichroic mirror and emission filters and finally detected with a photomultiplier tube. We used two types of tLCDs; one was a plain cell tLCD (tLCD-P) with homogeneously aligned LC molecules functioning as an applied voltage-dependent variable wave plate. The other was a 24-piece divided phase mask (tLCD-24) that enabled production of a spiral-phase distribution with 24 steps around the centre of the beam, as shown in the lower row of Fig. 6. For STED microscopy, the focus of an optical vortex should have a steep central dark spot, leading to a higher spatial resolution as the size of the centre of the dark spot is reduced [2]. As described in a previous section, to focus an optical vortex, the handedness of the circular polarization was chosen so as to create a fine dark spot at the focus [7]. However, for the case of a circularly polarized optical vortex at the position of the tLCDs, a difference in phase shifts between the s- and p-polarized light generated in the
231
232
Microscopy, 2015, Vol. 64, No. 4
and TP-STED images in Fig. 7b were averages of four and eight acquired images, respectively. After applying the STED light, the fluorescent bead image that had exhibited an isotropic shape in the focal plane became elliptical as the size itself became smaller. We estimated that the FWHM values were 322 and 173 nm for the averaged fluorescence intensity profiles acquired with conventional TPLSM and our TP-STED, respectively. To estimate the PSFs and the spatial resolution of the microscopy more accurately, measurements of the fluorescent images of smaller-sized beads would be required. However, the sensitivity of our conventional TPLSM system was insufficient to clearly visualize such small fluorescent beads, due to weak fluorescence. Finally, we demonstrated that our method was applicable to biological specimens by observing microtubules in fixed COS-7 cells that were immunostained with fluorescent dye-marked antibodies. ATTO 425 was chosen as the fluorescent dye, because the wavelength of our STED light hardly overlapped with its absorption band. The excitation laser power at the output of the microscope objective was 3.4 mW for both imaging. The STED laser power at the same position was 72 mW. The pixel size of both images was 41.4 × 41.4 nm, and the pixel dwell time was 10 μs. As shown in Fig. 8, TP-STED microscopy enabled us to visualize the fine network structures of microtubules more clearly than conventional TPLSM.
Summary and outlook As described in the previous section, by adding tLCDs and a CW laser to our conventional TPLSM system, we developed
a new TP-STED microscopy system with the following useful properties. First, by adjusting the voltage applied to a tLCD-P, we compensated for various phase shifts in the optical path to acquire a fine doughnut-shaped PSF and confirmed that this optical vortex, which had a 24-part helical phase distribution of 2π created by a tLCD-24, was sufficient for STED microscopy. Second, by modifying a conventional TPLSM system by the addition of tLCDs and a depletion laser light source, we developed a TP-STED microscopy system. In principle, this methodology is applicable not only for TPLSM but also for other types of laser scanning microscopies. Therefore, tLCD is expected to achieve super-resolution microscopy more readily. Third, our TP-STED microscopy system was employable at different wavelengths of STED laser light by only tuning the voltages applied to tLCDs. We confirmed that the same tLCDs enabled both 473 and 577 nm laser beams to be converted into optical vortices [9]. One feature of TPLSM is that various fluorescent dyes can be excited simultaneously using a single wavelength NIR light [11,21]. Thus, combined with a wavelength-tunable light source for STED processes, this methodology might allow for super-resolution microscopy to visualize multiple components in living specimens using a variety of fluorophores. Finally, the other type of tLCDs enables to convert a Gaussian beam into a vector beam, which has an inhomogeneous polarization distribution on the beam cross-sections [22,23]. For example, a radially polarized beam, classified as one of the vector beams, is known to generate a strong axial component of the electric field at the focus and has been applied to the molecular
Downloaded from http://jmicro.oxfordjournals.org/ by guest on November 16, 2015
Fig. 7. (a) Merged fluorescence image of green fluorescent beads with diameters of 0.10 μm (green) and orange fluorescent beads with diameters of 0.17 μm (magenta) placed on the same cover slip. (b) Comparisons of the TPLSM and TP-STED images of a green fluorescent bead with a diameter of 0.10 μm. The lower panels show the averaged fluorescence intensity profiles of 10 beads across the intensity centre along the red dashed lines in the fluorescent images. The inlet length value indicates the full width at half maximum.
Microscopy, 2015, Vol. 64, No. 4
233
orientation imaging [24]. In most STED microscopies, a Gaussian beam and an optical vortex with a topological charge of one are used as an excitation light and a STED light, respectively. These two beams have circular polarization at the focus, so that acquired STED images have no information about molecular orientations. On the other hand, a super-resolution molecular orientation imaging using STED methodology has been demonstrated using an azimuthally polarized STED light although only the lateral orientation was determined [25]. If a radially polarized excitation beam and a STED light possessing strong axial electric field at the focus are also used, super-resolution microscopy acquiring three-dimensional molecular orientations can be achieved. This methodology is expected to visualize the detailed nanostructure of the micro-domain in the plasma membrane and may be feasible by our tLCD technologies. The spatial resolution of our present TP-STED microscopy system was, however, estimated to be larger than 100 nm, although the spatial resolution of the TP-STED systems in previous studies were reported to be smaller than 100 nm [16–19]. The reason might be that we used a CW laser as the STED light source; the STED efficiency was reported to be superior under pulse-type STED microscopy [18,19]. However, Willig et al. [26] demonstrated CW STED microscopy visualized nanostructures
