Nano-imaging of specimens in liquid conditions is needed in various applications such as analysis of colloidal solutions and observation of microcrystals during growth and self-assembly. In the case of biological specimens, high resolution imaging is crucial for a deeper understanding of cell functions. Fluorescence microscopy has been widely used to analyze the dynamic behavior of cellular components in living cells, since stained molecules of interest in the specimens can be imaged with high contrast.1, 2 However, many structures in cells are too small to be resolved with a standard optical microscope because of the diffraction limit of light.
We developed a direct, electron-beam excitation assisted, optical microscope (D-EXA microscope) that combines a scanning electron microscope (SEM) with a fluorescence microscope.3, 4 This super-resolution fluorescence microscope has spatial resolution of a few tens of nanometers and can image the dynamic behavior of nanoparticles in liquid conditions. Samples of fluorescent particles or fluorescently labeled biological specimens are placed on a thin film that functions as an electron transparent window. A focused electron beam irradiates the sample from below to directly excite luminescence, which is referred to as cathodoluminescence.5–7 This yields higher spatial resolution than normal light excitation, because the electron beam can be focused in a smaller region than that of a light beam even if some of the electrons are scattered in the film. Specimens can be observed in their native state in a liquid environment because the film separates the environment of the specimen from the vacuum in the microscope.
Figure 1. Principle of the D-EXA microscope.
The principle of direct excitation with an electron beam in the D-EXA microscope is shown in Figure 1. We can focus this beam to an area of approximately 2nm in diameter in vacuum,7 but the focus spot will broaden to a few tens of nanometers due to electron scattering in the film and specimens. We estimated the amount of scattering with Monte-Carlo simulation. The direction of the electron beam is controlled by magnetic or electric fields along the beam path. To create a real-time two-dimensional image, we modulated these fields to scan the beam over the surface of the specimen (no mechanical moving parts are necessary).
The thin film separates the samples from the vacuum in the SEM, which allows us to keep specimens in a liquid or air environment and observe their dynamic behavior under various conditions. In biological applications, the D-EXA microscope allows for the analysis of cell functions at the nanometer-scale under physiological conditions.
Figure 2. Prototype of the D-EXA microscope. The electron beam from the scanning electron microscope irradiates the specimens from below. The fluorescent emission is viewed through the fluorescence microscope and recorded by the photomultiplier (PMT).
We designed and built a prototype D-EXA microscope (see Figure 2). The beam from the SEM irradiates the specimens from below and the excited emission is collected above in the fluorescence microscope.3, 4 A photo-multiplier (PMT) records the signal.
To evaluate the resolution of the D-EXA microscope, we observed dried zinc oxide (ZnO) nanoparticles under atmospheric pressure.8, 9 We first placed the 100nm-sized ZnO nanoparticles in ultra-pure water on the thin film and dried them at room temperature. We then imaged the nanoparticles with both a conventional epi-fluorescence microscope and the D-EXA microscope, as shown in Figure 3(a) and 3(b), respectively. Both images represent the same area of 9.7μm × 9.7μm, but the ZnO nanoparticles are resolved more clearly in Figure 3(b). These results show that the D-EXA microscope has a higher spatial resolution than that of the epi-fluorescence microscope.
Figure 3. Luminescence images of 100nm zinc oxide (ZnO) nanoparticles acquired with (a) conventional epi-fluorescence microscope and (b) the D-EXA microscope. (c) Isolated ZnO nanoparticles imaged with the D-EXA microscope. (d)-(f) Line profiles of the regions between the arrows in the corresponding images above. The light intensity is shown in arbitrary units (a.u.).
Figure 3(d) and (e) represent the line-profiles of the regions between the arrows in Figure 3(a) and (b), respectively. The data indicates that the D-EXA microscope is able to distinguish each ZnO nanoparticle, whereas the epi-fluorescence microscope cannot. This is because the spatial resolution of the epi-fluorescence microscope is limited to approximately 200nm due to the diffraction limit of light. In Figure 3(c), we show the luminescence image of isolated ZnO nanoparticles acquired with the D-EXA microscope. From the corresponding line profile depicted in Figure 3(f), we find that the full width at half maximum for the D-EXA microscope is 100nm.
Figure 4. Observations of living cells using the D-EXA microscope. (a) A luminescence image of living Chinese hamster ovary (CHO) cells in culture solution without any treatments. The intracellular granules indicated with arrows are observed as white spots and the cell membranes are observed as light-gray contrast against the dark-gray background. (b) A phase contrast microscope image of the living cells, representing the same area as in (a).
In a separate experiment, we observed living Chinese hamster ovary (CHO) cells with the D-EXA microscope to confirm the potential for live cell imaging. These were MARCO-expressing cells, where MARCO refers to a specific macrophage receptor. To keep the cells alive during imaging, we placed them in culture solution without any treatments, such as fixation or drying. We compared the luminescence image of the cells acquired with the D-EXA microscope in Figure 4(a) to a phase contrast microscope image in Figure 4(b). Several features are recognizable, such as the shape of each cell and some bright spots inside cells. This demonstrates that the D-EXA microscope is a useful tool for observation of living biological cells in physiological conditions.
In future work, we need to evaluate the damage to the specimen by electron beam irradiation. We can minimize this damage, as well as improve the spatial resolution, by optimizing the acceleration voltage of the electron beam and by adjusting the film thickness and composition.
In conclusion, the D-EXA microscope is a technique with high spatial resolution beyond the diffraction limit of light. We have shown that it has a spatial resolution of better than 100nm and that it can provide luminescence imaging of living cells in culture solution by time lapse imaging. We believe that the D-EXA microscope is a promising tool and opens up new applications for life sciences as well as other fields.
The authors are grateful for the support of CREST, Japan Science and Technologies Agency.
Department of Mechanical Engineering
Johoku, Hamamatsu, Japan
Yoshimasa Kawata gained a PhD degree in applied physics from Osaka University in 1992. He worked at Osaka University as an assistant professor and moved to Shizuoka University in 1997, where he is currently chairman of the Department of Mechanical Engineering. He was selected as a distinguished researcher in 2011.
Division of Global Research Leaders
1. J. Lippincott-Schwartz, E. Snapp, A. Kenworthy, Studying protein dynamics in living cells, Nat. Rev. Mol. Cell Biol. 2, p. 444-456, 2001.
2. R. Y. Tsien, Imagining imaging's future, Nat. Rev. Mol. Cell Biol. 5, p. SS16-SS21, 2003.
3. W. Inami, K. Nakajima, A. Miyakawa, Y. Kawata, Electron beam excitation assisted optical microscope with ultra-high resolution, Opt. Express 18, p. 12897-12902, 2010.
4. Y. Nawa, W. Inami, A. Chiba, A. Ono, A. Miyakawa, Y. Kawata, S. Lin, S. Terakawa, Dynamic and high-resolution live cell imaging by direct electron beam excitation, Opt. Express 20, p. 5629-5635, 2012.
5. U. Fano, A theory on cathode luminescence, Phys. Rev. 5, p. 544-553, 1940.
6. J. Menniger, U. Jahn, O. Brandt, H. Yang, K. Ploog, Identification of optical transitions in cubic and hexagonal GaN by spatially resolved cathodoluminescence, Phys. Rev. B Condens. Matter 53, p. 1881-1885, 1996.
7. F. J. G. de Abajo, Optical excitations in electron microscopy, Rev. Mod. Phys. 82, p. 209-275, 1940.
8. L. N. Dem'yanets, L. E. Li, T. G. Uvarova, Hydrothermal synthesis and cathodoluminescence of ZnO crystalline powders and coatings, J. Cryst. Growth 287, p. 23-27, 2006.
9. K. Senthilkumar, O. Senthilkumar, K. Yamauchi, M. Sato, S. Morito, T. Ohba, M. Nakamura, Y. Fujita, Preparation of ZnO nanoparticles for bio-imaging applications, Phys. Status Solidi B 246, p. 885-888, 2009.