High-resolution microscopy for biological specimens was performed using cathodoluminescence (CL) of Y2O3:Eu, Zn nanophosphors, which have high CL intensity due to the incorporation of Zn. The intensity of Y2O3:Eu nanophosphors at low acceleration voltage (3 kV) was increased by adding Zn. The CL intensity was high enough for imaging even with a phosphor size as small as about 30 nm. The results show the possibility of using CL microscopy for biological specimens at single-protein-scale resolution. CL imaging of HeLa cells containing laser-ablated Y2O3:Eu, Zn nanophosphors achieved a spatial resolution of a few tens of nanometers. Y2O3:Eu, Zn nanophosphors in HeLa cells were also imaged with 254 nm ultraviolet light excitation. The results suggest that correlative microscopy using CL, secondary electrons and fluorescence imaging could enable multi-scale investigation of molecular localization from the nanoscale to the microscale.
© 2013 Optical Society of America
A novel imaging technique for imaging proteins in cells with both molecular specificity and nanometer-order spatial resolution is highly desirable for revealing cellular functions with distributions of different types of biological molecules, because conventional cellular imaging methods, such as light microscopy (LM) and electron microscopy (EM), have functional limitations. LM reveals the spatiotemporal distributions of several types of proteins with fluorescent probes; however, it is difficult to obtain detailed information about the localization of proteins due to the diffraction limit of light. EM is able to observe at a higher spatial resolution than LM because the spot size of the focused electron beam is on the order of nanometers. EM allows us to observe a molecular species at nanoscale resolution through immuno-stained molecules with gold nanoparticles (GNPs) [1, 2]. In addition, by using GNPs of different sizes, more than one kind of molecular species can be identified [3–5]. However, size discrimination of the different GNPs is difficult in the case of complex distributions of proteins because the obtained image is monochrome due to the detection of electrons, and discrimination becomes even more difficult if three or more types of GNPs are used.
To overcome these functional limitations, many researchers have developed new biological imaging methods. In the LM field, nowadays, many kinds of super-resolution microscopy that are capable of imaging beyond the diffraction limit have attracted attention [6, 7]. Moreover, correlative light and electron microscopy (CLEM) was developed to bridge the functional gap between LM and EM [8, 9]. In CLEM, LM allows fast screening of many kinds of molecular distributions over a large sample area, and EM gives detailed information in the region of interest. Cathodoluminescence (CL) microscopy is also a promising new technique that can solve the above problems. CL is the light emission from a material induced by accelerated electrons, and its luminescence color depends on the material . CL microscopy enables multi-color, nanoscale imaging of protein distributions at resolutions beyond the diffraction limit of light because it is based on electron beam excitation . In addition, EM offers cellular structural information while a CL image is taken at the same position of the sample. Owing to these attractive functions, this novel imaging technique, having both molecular specificity and nanoscale resolution, is expected to be useful for observation of biological specimens.
To realize biological CL imaging, there is a great demand for CL probes with high intensity, nanometer size, high resistance against electron beam excitation, and different colors. Several probes have been investigated for CL imaging of biological tissue and cells, such as green fluorescent protein , quantum dots , ZnO nanoparticles , organic molecules , and rare earth-doped nanophosphors (RE-NPs) . In particular, oxide-based RE-NPs have high CL intensity and high resistance against electron-beam excitation , and the emission wavelength is easily selected by changing the activator ions (e.g., Eu (red), Tb (green), and Tm (blue)) . The spectral bandwidth from the activator ions is quite narrow. Thus, our group has focused on oxide-based RE-NPs (particularly Y2O3:Eu, Y2O3:Tb, and Y2O3:Tm) for biological CL imaging and has demonstrated multi-color CL imaging of cells . As an additional advantage of RE-NPs, they can be excited by light, not just an electron beam. EM and fluorescence imaging using these phosphors allows multi-scale investigation of molecular localization from the nanoscale to the microscale.
Although Y2O3-based RE-NPs have many advantages, two fundamental problems still remain when applied to biological CL imaging. First, the size of each phosphor has to be small enough to image protein distributions because the spatial resolution depends on the size of the phosphors, due to the nanometer-order high spatial resolution of CL microscopy, and the phosphors should also be well-dispersed. Second, the intensity of the phosphors is not high enough for biological CL microscopy with nanophosphors as small as a few tens of nanometers (the typical size of a protein is a few nanometers to a few tens of nanometers) because the CL intensity of an RE-NP is proportional to its volume.
This paper describes improvement of the CL intensity of Y2O3:Eu nanophosphors by adding Zn, and CL imaging of a single nanophosphor (size about 30 nm) for practical use as a biological CL imaging probe. Well-dispersed small nanophosphors were synthesized by laser ablation and were introduced into HeLa cells through endocytosis. We performed biological CL imaging using laser-ablated Zn-doped Y2O3:Eu nanophosphors, which is the first demonstration of CL microscopy at high spatial resolution. We investigated the effect of acceleration voltage on the spatial resolution. Furthermore, for correlative microscopy, nanophosphors in HeLa cells were also imaged under ultraviolet (UV) light excitation.
2. Second electron microscopy - cathodoluminescence system
We observed the CL spectra and CL images by using a field-emission scanning electron microscope (FE-SEM; JEOL JSM-6500F) equipped with a CL measurement unit (HORIBA). The CL microscope is illustrated in Fig. 1(a). CL emission induced by a focused electron beam was collected using an ellipsoidal mirror and detected through an optical fiber bundle and a spectrometer. CL spectra were detected by a charge-coupled device (CCD), and a photomultiplier tube (PMT) was used for CL imaging. The spectral resolution of CL spectra was determined by the entrance slit of the spectrometer; all CL images were taken with a 2 mm slit width to obtain a larger CL signal. The center wavelength for CL imaging was selected by the angle of the grating. The full width at half-maximum of the measured spectral bandwidth was about 35 nm when the slit width was 2 mm and the center wavelength was 614 nm. Red intense peaks of Y2O3:Eu ((Y0.95Eu0.05)2O3) on a P-doped Si substrate were observed [Fig. 1(b)). The phosphor samples were prepared by a sol-gel method detailed in our previous paper . SEM and CL images were constructed by raster scanning of a focused electron beam [Fig. 1(c), 1(d)). Both images showed similar phosphor distributions.
3. Improvement of CL intensity of Y2O3:Eu nanophosphors with doping Zn
Zn-doped Y2O3:Eu nanophosphors were synthesized to improve the CL intensity of the nanophosphors. The surface of the phosphors is negatively charged by irradiation of electrons because of their low conductivity, and these negative charges hamper penetration of irradiated electrons. Thus, a practical level of CL intensity from the phosphors cannot be obtained . Zn2+in the Y2O3lattice plays the role of an electron conductor, which prevents charging of the phosphors, and the electrons penetrate deeply into the phosphors. Since the penetrating electrons increase the excitation volume, increased CL intensity is expected [18, 19].
To investigate the effect of Zn doping on the CL intensity, nanophosphors with different Zn concentrations were synthesized ((Y0.95-XEu0.05ZnX)2O3, X = 0, 0.05, 0.15, and 0.30). Y2O3:Eu, Zn nanophosphors were prepared by a sol-gel process using metal nitrate and a gelling agent. Europium was selected as the activator ion of red phosphors. Yttrium nitrate, zinc nitrate, and europium nitrate were dissolved in water (0.06 mol/L) together with glutamic acid, and this solution was dried at 110 °C. After drying, the sample was annealed at 900 °C for 3 h in air to obtain Y2O3:Eu, Zn in the form of a white crystalline powder. The primary size of the Y2O3:Eu, Zn phosphors was determined by transmission electron microscopy (TEM) to be less than 100 nm. The size of aggregated phosphors was several tens to several hundred μm. Almost all of the nanophosphors were aggregated due to the annealing process at 900 °C.
In both Y2O3:Eu and Y2O3:Eu, Zn, the most intense red emission attributed to 5D0⇒7F2was observed. The f shell of the rare-earth ion is shielded from external electric fields due to 5s25p6shells existing outside the f shell . Thus, these CL spectra show similar shapes and quite narrow spectral bandwidth. This narrow spectral bandwidth permits us to distinguish each biological molecular distribution when different colors of RE-NPs are used simultaneously.
Figure 2shows the CL intensity as a function of the beam current using Y2O3:Eu nanophosphors having different Zn concentrations. Phosphors with a size of about 200 nm were selected. The more the Zn concentration was increased, the more the CL intensity of the phosphors increased. Remarkably high CL intensity was obtained at a high beam current, suggesting the possibility of using these Zn-doped phosphors for high-speed imaging in biological CL microscopy. However, the standard deviation of the CL intensity of the phosphors increased at higher Zn concentration. Although the reason for this may depend on the difference in crystallinity of each phosphor, the details are still under investigation.
Figure 3shows SEM and CL images of Y2O3:Eu, Zn ((Y0.8Eu0.05Zn0.15)2O3) nanophosphors. The CL emission intensity from the Y2O3:Eu, Zn nanophosphors was high enough for CL imaging, even though the diameter of the nanophosphors was as small as 30 nm. This result implies that the spatial resolution of CL microscopy can reach the single-protein level required for biological specimens.
4. Preparation of small Y2O3:Eu, Zn nanophosphors with using laser ablation method
An FE-SEM-CL system has nanoscale spatial resolution (a few nanometers) due to the small spot size of the electron beam. To obtain high-spatial-resolution CL images using this system, the size of the individual phosphors has to be small. Laser ablation in liquid is a well-known method for producing small nanopart by means of a transient plasma plume in the laser-focused region [20–23]. By using laser ablation, aggregated Y2O3:Eu, Zn nanophosphors were dispersed in water, and the size of the phosphors was reduced. A nanosecond Nd:YAG laser beam (532 nm, 10 Hz, 20 mJ/pulse) was focused into the phosphor solution by a lens (f = 100 mm). The solution was stirred with a magnetic stirrer while being treated with laser ablation for 3 hours.
Figures 4(a)and 4(b)show SEM and TEM images of Y2O3:Eu, Zn ((Y0.8Eu0.05Zn0.15)2O3) phosphors before laser ablation, and Figs. 4(c)and 4(d)show SEM and TEM images after laser ablation. The figures show that the nanophosphors were well dispersed, and the size was from a few nanometers to a few hundred nanometers. Furthermore, the particles became round in shape. The reason for this is thought to be related to the growth process of the phosphors in the plasma plume at the laser-focused region . The plasma plume creates high-temperature and high-pressure conditions, and nucleation and growth take place. This process performed under liquid conditions may be important for synthesizing spherical phosphors. From the above results, small, well-dispersed nanophosphors ranging from a few nanometers to 150 nm were obtained by the laser ablation method.
5. Biological CL imaging using HeLa cells and laser-ablated Y2O3:Eu, Zn
CL imaging of HeLa cells containing laser-ablated Y2O3:Eu, Zn ((Y0.8Eu0.05Zn0.15)2O3) injected by an endocytotic process was carried out. Specimens were prepared by treating the cells using typical biological sample preparation protocols for TEM observation, such as fixation, dehydration, and embedding in epoxy resin . Thin sections of the cell specimens sliced to a thickness of 100 nm were placed on a P-doped Si substrate, the epoxy resin was removed with a saturated KOH/ethanol solution to expose the structure of the cells, and a 10 nm gold layer was sputtered on the sliced cell specimens.
SEM and CL images of Y2O3:Eu, Zn ((Y0.8Eu0.05Zn0.15)2O3) in a HeLa cell are shown in Figs. 5(a)-5(c). Figure 5(a)is a low-magnification image, and Fig. 5(b)is a high-magnification image corresponding to the area indicated by the red rectangle in Fig. 5(a). The SEM images show an endocytotic vesicle including two nanophosphors. Figure 5(c)is a CL image of the phosphors observed at a wavelength of 614 nm in the same region as in (b). The nanophosphors in the CL image were observed at almost the same spatial resolution as those in the SEM image. The size of the phosphors was about 200 nm to 300 nm (these phosphors may have been aggregations of laser-ablated small nanophosphors), and the gap between the two phosphors was about 100 nm. Thus, the spatial resolution of this CL imaging technique was at least a few tens of nanometers because the gap was clearly resolved.
The spatial resolution of this CL microscopy technique also depended on the acceleration voltage of the electron beam and the thickness of the sample. Figure 6shows CL images observed with various acceleration voltages (3 kV, 5 kV, and 10 kV). Line profiles (A–B) of the four images are plotted in Fig. 6(e)to compare the resolution of each image. The beam current was set to 53–66 pA. An increase in the scattering distance of the electrons due to the high acceleration voltage indirectly induces undesired emission from the phosphors even though the electron beam scans outside the nanophosphors . The spatial resolution of the CL image was highest at 3 kV and was comparable to that of the SEM image. The acceleration voltage should be kept low for biological CL imaging. The relation between the spatial resolution of the CL image and the thickness of the sample was also observed using samples with different thicknesses (data not shown here). The spatial resolution of a thin sample was higher than that of a thick sample because the excitation volume of nanophosphors by the electron beam increased with the sample thickness. Therefore, to realize CL imaging with single-protein-level spatial resolution, it is necessary to use small nanophosphors, a low acceleration voltage, and thin samples. The TEM-CL method described here is another candidate technique in which electron scattering from the specimen is reduced due to the absence of a substrate and a higher acceleration voltage than SEM [26, 27]. TEM-CL enables a smaller scattering volume in thin samples compared with SEM, and the smaller scattering volume reduces undesired CL from outside the phosphors. Furthermore, using TEM-CL allows us to obtain both TEM images of cellular components and CL images of the distribution of nanophosphors in the cells.
6. Fluorescence imaging using HeLa cells and laser-ablated Y2O3:Eu, Zn
Laser-ablated Y2O3:Eu, Zn nanophosphors were also observed with optical fluorescence microscopy by UV light excitation [Fig. 7(a)]. The Y2O3:Eu, Zn nanophosphors inside HeLa cells were imaged using the 254 nm spectral line of a mercury lamp. The cells were fixed with 4% paraformaldehyde after injection of the laser-ablated nanophosphors via endocytosis. Figure 7shows transmission (b), fluorescence (c), and combined (d) images of nanophosphors contained in HeLa cells. This observation technique will allow us to image the distribution of nanophosphors under fully hydrated conditions before CL observation. It will also allow us to confirm the conditions (e.g., well-stained or not) of specimens before the time-consuming preparation of sliced samples, and this is also useful for TEM-CL observation of thin sections. This technique might be a promising new approach for CLEM, which can be called correlative light and cathodoluminescence microscopy. RE-NPs have the potential to allow multi-scale imaging of molecular localization from the nanoscale to the microscale.
In conclusion, highly luminescent nanophosphors for biological CL imaging were synthesized by adding Zn to Y2O3:Eu. The laser ablation method provided well-dispersed small nanophosphors (a few nanometers to a few hundred nanometers). We demonstrated biological cathodoluminescence (CL) imaging of RE-NPs within HeLa cells at a high spatial resolution of a few tens of nanometers. RE-NPs can be excited by UV light, and fluorescence imaging is possible. Thus, it will be possible to realize CLEM using CL, SEM, and fluorescence images, which will allow multiscale imaging from the nanoscale to the microscale. If we employ upconversion nanophosphors (UC-NPs) such as Y2O3: Er [28, 29] and Y2O3: Yb, Er  simply by changing the rare-earth ions, nanoscale (CL) to millimeter-scale (near infrared (NIR)) imaging might be possible because excitation and emission of UC-NPs both occur in the NIR region. In addition, with CL, the molecular species can be distinguished even if the cells are in wet conditions [31, 32]. For the reasons described above, this technique will be a superior imaging technique for biological specimens, allowing protein species to be distinguished at high spatial resolution by immunostaining. CL imaging with RE-NPs will also be useful for TEM-CL.
This research was partially supported by KAKENHIfrom the Ministry of Education, Culture, Sports, Science and Technology, Japan, and by the Tateisi Science and Technology Foundation, Kyoto, Japanand the Nakatani Foundation, Tokyo, Japan.
References and links
3. B. L. Wang and L. I. Larsson, “Simultaneous demonstration of multiple antigens by indirect immunofluorescence or immunogold staining. Novel light and electron microscopical double and triple staining method employing primary antibodies from the same species,” Histochemistry 83(1), 47–56(1985). [CrossRef] [PubMed]
4. R. S. Petralia, J. A. Esteban, Y. X. Wang, J. G. Partridge, H. M. Zhao, R. J. Wenthold, and R. Malinow, “Selective acquisition of AMPA receptors over postnatal development suggests a molecular basis for silent synapses,” Nat. Neurosci. 2(1), 31–36(1999). [CrossRef] [PubMed]
5. H. J. Geuze, J. W. Slot, P. A. van der Ley, R. C. Scheffer, and J. M. Griffith, “Use of colloidal gold particles in double-labeling immunoelectron microscopy of ultrathin frozen tissue sections,” J. Cell Biol. 89(3), 653–665(1981). [CrossRef] [PubMed]
9. T. M. Svitkina, E. A. Bulanova, O. Y. Chaga, D. M. Vignjevic, S. Kojima, J. M. Vasiliev, and G. G. Borisy, “Mechanism of filopodia initiation by reorganization of a dendritic network,” J. Cell Biol. 160(3), 409–421(2003). [CrossRef] [PubMed]
10. B. G. Yacobi and D. B. Holt, Cathodoluminescence Microscopy of Inorganic Solids(Plenum, 1990).
11. H. Niioka, T. Furukawa, M. Ichimiya, M. Ashida, T. Araki, and M. Hashimoto, “Multicolor cathodoluminescence microscopy for biological imaging with nanophosphors,” Appl. Phys. Express 4(11), 112402(2011). [CrossRef]
12. P. J. Fisher, W. S. Wessels, A. B. Dietz, and F. G. Prendergast, “Enhanced biological cathodoluminescence,” Opt. Commun. 281(7), 1901–1908(2008). [CrossRef]
13. Y. Nawa, W. Inami, A. Chiba, A. Ono, A. Miyakawa, Y. Kawata, S. Lin, and S. Terakawa, “Dynamic and high-resolution live cell imaging by direct electron beam excitation,” Opt. Express 20(5), 5629–5635(2012). [CrossRef] [PubMed]
15. P. J. Fisher, W. S. Wessels, A. B. Dietz, and F. G. Prendergast, “Immuno-fluorescence scanning electron microscopy of biological cells,” Opt. Commun. 18(5), 8–13(2010).
16. W. M. Yen, S. Shionoya, and H. Yamamoto, Phosphor Handbook(CRC, 2007).
17. J. Hao, S. A. Studenikin, and M. J. Cocivera, “Blue, green and red cathodoluminescence of Y2O3phosphor films prepared by spray pyrolysis,” Lumin. 93(4), 313–319(2001). [CrossRef]
18. T. Nagata, T. Nakayama, and H. Murakami, “Low voltage cathodoluminescent properties of Y2O3-based nano-particle phosphors,” SID’ 07 Digest (2007), 1328-1331(2007).
19. S. Sakuma, H. Kominami, Y. Neo, T. Aoki, Y. Nakanishi, and H. Mimura, “Effect of La and Zn addition on Y2O3:Eu phosphors,” Appl. Surf. Sci. 244(1-4), 458–460(2005). [CrossRef]
20. F. Mafuné, J. Kohno, Y. Takeda, T. Kondow, and H. Sawabe, “Formation and size control of silver nanoparticles by laser ablation in aqueous solution,” J. Phys. Chem. B 104(39), 9111–9117(2000). [CrossRef]
21. A. Murali, A. Barve, V. J. Leppert, S. H. Risbud, I. M. Kennedy, and H. W. H. Lee, “Synthesis and characterization of indium oxide nanoparticles,” Nano Lett. 1(6), 287–289(2001). [CrossRef]
22. G. Ledoux, D. Amans, C. Dujardin, and K. Masenelli-Varlot, “Facile and rapid synthesis of highly luminescent nanoparticles via pulsed laser ablation in liquid,” Nanotechnology 20(44), 445605(2009). [CrossRef] [PubMed]
23. H. Zeng, S. Yang, and W. Cai, “Reshaping formation and luminescence evolution of ZnO quantum dots by laser-induced fragmentation in liquid,” J. Phys. Chem. C 115(12), 5038–5043(2011). [CrossRef]
25. D. C. Joy, Monte Carlo Modeling for Electron Microscopy and Microanalysis(Oxford, 1995).
26. N. Yamamoto, K. Araya, and F. García de Abajo, “Photon emission from silver particles induced by a high-energy electron beam,” J. Phys. Rev. B 64(20), 205419(2001). [CrossRef]
28. S. Chandra, F. L. Deepak, J. B. Gruber, and D. K. Sardar, “Synthesis, morphology, and optical characterization of nanocrystalline Er3+:Y2O3,” J. Phys. Chem. C 114(2), 874–880(2010). [CrossRef]
29. V. Singh, V. K. Rai, I. Ledoux-Rak, S. Watanabe, T. K. G. Rao, J. F. D. Chucaci, L. Badie, F. Pelle, and S. Ivanova, “NIR to visible up-conversion, infrared luminescence, thermoluminescence and defect centres in Y2O3: Er phosphor,” J. Phys. D Appl. Phys. 42(6), 065104(2009). [CrossRef]
30. S. F. Lim, R. Riehn, W. S. Ryu, N. Khanarian, C. K. Tung, D. Tank, and R. H. Austin, “In vivo and scanning electron microscopy imaging of up-converting nanophosphors in Caenorhabditis elegans,” Nano Lett. 6(2), 169–174(2006). [CrossRef] [PubMed]
31. S. Thiberge, A. Nechushtan, D. Sprinzak, O. Gileadi, V. Behar, O. Zik, Y. Chowers, S. Michaeli, J. Schlessinger, and E. Moses, “Scanning electron microscopy of cells and tissues under fully hydrated conditions,” Proc. Natl. Acad. Sci. U.S.A. 101(10), 3346–3351(2004). [CrossRef] [PubMed]