TY - JOUR
AU1 - Nakakoshi,, Masamichi
AU2 - Nishioka,, Hideo
AU3 - Katayama,, Eisaku
AB - Abstract Aqueous uranyl acetate has been extensively used as a superb staining reagent for transmission electron microscopy of biological materials. However, recent regulation of nuclear fuel material severely restricts its use even for purely scientific purposes. Since uranyl salts are hazardous due to biological toxicity and remaining radioactivity, development of safe and non-radioactive substitutes is greatly anticipated. We examined two lanthanide salts, samarium triacetate and gadolinium triacetate, and found that 1–10% solution of these reagents was safe but still possess excellent capability for staining thin sections of plastic-embedded materials of animal and plant origin. Although post-fixation with osmium tetroxide was essential for high-contrast staining, post-staining with lead citrate could be eliminated if a slow-scan CCD camera is available for observation. These lanthanide salts can also be utilized as good negative-staining reagents to study supramolecular architecture of biological macromolecules. They were not as effective as a fixative of protein assembly, reflecting the non-hazardous nature of the reagents. transmission electron microscopy, staining reagent, negative staining, uranyl acetate substitute, samarium triacetate, gadolinium triacetate Introduction Since the introduction of heavy metal salts as versatile staining reagents for transmission electron microscopy of biological materials [1,2], uranyl acetate has been extensively used as a superb stain not only for thin sections of plastic-embedded tissues and cells, but also for high-contrast negative stains for biological macromolecules [3,4]. Though majority of radioactive nuclides are depleted from unanyl salts used for electron microscopy, they are still categorized nominally into nuclear fuels and their use is severely restricted (http://www.hpa.org.uk/Topics/Radiation/UnderstandingRadiation/FrequentlyAskedQuestions/DepletedUranium/#3) even for purely scientific purposes. Scientists or technical staff in newly organized laboratories hardly share the benefit of this useful reagent, because even a purchase of small amount of reagents requires a tedious process to obtain government authorization. Though not as much as such a case, even already established microscopy laboratories share the difficulty of the safe storage of reagents and the inability to dispose after using them. Introduction of a safe substitute that could match the high performance of uranyl salts is an urgent global requirement. Similar attempts have already been made by several laboratories, mostly including Japanese authors. They have examined the performance of platinum-blue [5], oolong tea extract [6] and hafnium chloride [7], some of which are commercially available as new staining regents. These studies were made mostly for plastic-embedded thin sections and few authors mentioned their use as negative-staining reagents. Inspired by the chemical properties of uranyl acetate, we tested various heavy metal salts and found that triacetate salts of gadolinium and samarium could be the right candidates for our present purpose. Brenner and Horne [3] were one of the first to introduce the negative-staining technique to explore the gross structure of macromolecular assembly. They showed negatively contrasted images of tobacco mosaic virus and some other materials embedded in a layer of 2% sodium phosphotungstate. Ever since van Bruggen et al. [8] reported uranyl diacetate, UO2(CH3COO)2, as a high-contrast medium for negative staining, various uranyl salts have been frequently used as a useful stain to observe inter- or intramolecular structural details of biological macromolecular complex by transmission electron microscopy. Though several more reagents were introduced for negative staining since then [9], nothing comparable to uranyl salts in terms of the image contrast is yet been reported. We thus tested the performance of the new heavy metal salts as a negative-staining reagent, probing whether they could be used as a substitute for uranyl acetate. Lanthanum (La, atomic number 57), samarium (Sm, atomic number 62) and gadolinium (Gd, atomic number 64) are the members of the lanthanide group. Among the metal reagents evaluated for their applicability to negative staining, Bradley [9] noted that lanthanum triacetate, together with thorium nitrate or chloride, could certainly give negatively contrasted images. Lanthanum salt was not used until Leeson and Higgs [10] revisited the membrane permeability of lanthanum trichloride' to a different purpose of ‘intracellular staining’. Since then, there has been no application, to our knowledge, of lanthanum in electron microscopy. Electron-staining effect of the other lanthanides has not yet been reported. In this report, we show the performance of samarium and gadolinium salts as novel staining reagents for biological electron microscopy. Materials and methods Thin sections for the test observation were prepared from animal and plant tissues as follows. Mouse kidney, liver and hippocampus, and leaves of Spinacia oleracea were pre-fixed with 2.5% glutaraldehyde for 2 h, followed by 1 h post-fixation with 1% osmium tetroxide (OsO4), both in 0.1 M sodium cacodyrate buffer (pH 7.2). They were dehydrated through a graded series of ethanol and were embedded in Epon 812. Ultra-thin sections were cut with Leica Ultracut-S (Leica Co. Ltd) and directly mounted onto 400-mesh copper grids. They were stained on the droplet of either 2.5% samarium triacetate (Sm(CH3COO)3), 2.5% gadolinium triacetate Gd(CH3COO)3), 2.5% meglumine gadoterate (gadolinium 1,4,7-tris(carboxymethyl)-10-(2′-hydroxypropyl)- 1,4,7,10-tetraazacyclododecane) or 4% uranyl diacetate (UO2(CH3COO)2) for 20 min at room temperature and rinsed thrice with several drops of distilled water. Some of them were post-stained by Reynolds' lead citrate [11] for 20 min. Negative-staining effect of the new stains was examined with two kinds of macromolecular complex, amyloid fibril of β1–42 (Aβ1–42) and raw preparation of bacteriophage T4 (a kind gift from Dr F. Arisaka). Materials were directly adsorbed onto a polyvinyl formal supporting membrane on a 400-mesh copper grid, stained with either 2.5% samarium triacetate, 2.5% gadolinium triacetate or 4% uranyl diacetate, all dissolved in distilled water, and dried at room temperature, according to a conventional protocol. Capability of these reagents as a fixative prior to rotary shadowing [12] was tested using native thin-filament of smooth muscle [13]. Samples were subjected to observation under transmission electron microscope (JEM-1200EX and JEM-1400; JEOL Ltd) at an acceleration voltage of 120 kV with an objective aperture of 20 μm. Micrographs in Figs. 1 and 2 were taken with a 16-bit slow-scan CCD camera (1024×1024 pixels) at 5000× nominal magnification (i.e. 5.1 nm sampling unit per pixel) and 70 μm defocusing. Electron dose at the specimen position was 1300 pA cm−2. Negatively stained images in Fig. 3 were taken with a Fuji electron microscopic film at 20 000× nominal magnification with ∼1.5 μm defocusing. Electron dose was 470 pA cm−2. Negative films were scanned at 1200 dpi to give 1 nm per pixel. Fig. 1. Open in new tabDownload slide Images of plastic-embedded mouse kidney stained with metal acetates and meglumine gadoterate. Original 16-bit gray-scale images of metal-stained sections were taken under similar degree of beam convergence for illumination. Since the brightness levels of the clear background area of respective original images were not much different from each other, they were adjusted to the same level as an offset. Then, all the images were combined and linearly enhanced to the appropriate contrast to give final images in the gallery. (a) Images without post-staining, arranged from top to bottom, no metal-staining, samarium triacetate, gadolinium triacetate, meglumine gadoterate and uranyl acetate-staining. (b) Similar images but with lead post-staining. (c) Histogram of brightness distribution to compare the degree of staining with respective metal reagents. In each panel, shaded and hatched area correspond to those without and with post-staining, respectively, whereas filled area exhibits the distribution in non-stained image. Black, gray and white triangles along the ordinate indicate minimum, center and maximum density in a 16-bit gray-scale (65 536 levels), respectively. Relative stain intensity among cell organelles in lanthanide-stained sections was almost indistinguishable from that with uranyl acetate. Scale bar, 1 μm. Fig. 1. Open in new tabDownload slide Images of plastic-embedded mouse kidney stained with metal acetates and meglumine gadoterate. Original 16-bit gray-scale images of metal-stained sections were taken under similar degree of beam convergence for illumination. Since the brightness levels of the clear background area of respective original images were not much different from each other, they were adjusted to the same level as an offset. Then, all the images were combined and linearly enhanced to the appropriate contrast to give final images in the gallery. (a) Images without post-staining, arranged from top to bottom, no metal-staining, samarium triacetate, gadolinium triacetate, meglumine gadoterate and uranyl acetate-staining. (b) Similar images but with lead post-staining. (c) Histogram of brightness distribution to compare the degree of staining with respective metal reagents. In each panel, shaded and hatched area correspond to those without and with post-staining, respectively, whereas filled area exhibits the distribution in non-stained image. Black, gray and white triangles along the ordinate indicate minimum, center and maximum density in a 16-bit gray-scale (65 536 levels), respectively. Relative stain intensity among cell organelles in lanthanide-stained sections was almost indistinguishable from that with uranyl acetate. Scale bar, 1 μm. Fig. 2. Open in new tabDownload slide Metal-stained images of plant tissues. Thin sections of plastic-embedded leaf of S. oleracea were stained with metal acetates and meglumine gadoterate. (a) Images without post-staining, arranged from top to bottom, samarium triacetate, gadolinium triacetate, meglumine gadoterate and uranyl acetate-staining. (b) Similar images but with lead post-staining. Scale bar, 1 μm. Fig. 2. Open in new tabDownload slide Metal-stained images of plant tissues. Thin sections of plastic-embedded leaf of S. oleracea were stained with metal acetates and meglumine gadoterate. (a) Images without post-staining, arranged from top to bottom, samarium triacetate, gadolinium triacetate, meglumine gadoterate and uranyl acetate-staining. (b) Similar images but with lead post-staining. Scale bar, 1 μm. Fig. 3. Open in new tabDownload slide (a) Images of Aβ1–42 fibril negatively stained with (from left to right) 2.5% samarium acetate, 2.5% gadolinium acetate and 4% uranyl acetate. (b) Low- and high-magnification (insets) views of bacteriophage T4 exhibiting the arrangement of the T4 phage components; head, tail and tail fibers (arrowheads); comparable to authentic images (e.g. Ref. 11). Scale bars, 100 and 20 nm, respectively. Fig. 3. Open in new tabDownload slide (a) Images of Aβ1–42 fibril negatively stained with (from left to right) 2.5% samarium acetate, 2.5% gadolinium acetate and 4% uranyl acetate. (b) Low- and high-magnification (insets) views of bacteriophage T4 exhibiting the arrangement of the T4 phage components; head, tail and tail fibers (arrowheads); comparable to authentic images (e.g. Ref. 11). Scale bars, 100 and 20 nm, respectively. Photographic density distributions in the metal-stained images were evaluated by comparing their histograms taken in a 16-bit scale. Results Staining of thin sections with new heavy metal reagents Heavy metal staining in biological electron microscopy might arise from the electron-beam scattering effect of metal ions bound to various groups on biological materials. Uranium atom, one of the most efficient reagents to stain biological materials, has a strong electron-beam scattering effect due to its large atomic number and atomic radius. During the course of searching for the reagent with a similar property, we found that two lanthanide elements, Sm and Gd, have almost the same atomic radius as that of uranium. We also confirmed that the compounds including those elements are safe, not designated as toxic or hazardous substances under the law. Thus, we examined their physical properties, taking them as potentially right candidates of the new stain. While the solutions of SmCl3 and GdCl3, and Sm(OH)3 and Gd(OH)3 gave acidic and basic pH, respectively, their triacetate salts, Sm(CH3COO)3 and Gd(CH3COO)3, were both reasonably soluble in water, showing a neutral pH at room temperature. They did not form crystals when the solution dried up, suggesting that they might be compatible also as negative-staining reagents. The resultant images of animal (Fig. 1a and b) and plant tissues (Fig. 2) together with the histograms of the former (Fig. 1c) indicate that three metal reagents coupled with lead post-staining provide enough contrast almost comparable to that of uranyl acetate. It is notable that relative stain distribution among various organelles in the cell was hardly distinguishable from that of uranyl acetate (only the images of kidney are shown as an example of animal tissue, but the tendency was the same for the other tissues as well). This suggests that the binding mode of new metal reagents to cell components might be mostly ionic binding with negatively charged groups such as carboxyl or phosphate groups, in common with uranyl salts. Meglumine gadoterate is known as a highly inert contrast media for magnetic resonance imaging [14]. Its structure containing gadolinium reminds us that it might be another candidate of electron stain. Though the performance seemed weaker than the others, it gave a certain degree of contrast to the sections, especially with lead post-staining (Fig. 1b). Encouraged by the successful results, we then made further trial of single-reagent staining without post-staining. Figures 1a and 2a exhibit some of such images. Though the contrast was slightly less than that with post-staining, it is apparent that they are still well acceptable, especially when the pictures were taken with a slow-scan CCD camera, where one can easily enhance the contrast by a digital image processing (subtraction of dark noise and flat-fielding, i.e. normalization of sensitivity of each pixel'). Osmium tetroxide is another hazardous reagent which seems to be mandatory in biological electron microscopy. We also examined whether the possibility of osmium post-fixation could be skipped with the new heavy metal salts for staining. Unfortunately, however, we did not obtain enough stain density without post-fixation and ultimately recognized the indispensable role of osmium treatment as a prerequisite for heavy metal staining. Negative staining with new heavy metal reagents We then examined the performance of these reagents as negative stains to observe the architecture of biological macromolecules. Since the materials are, usually, not prefixed for negative staining, it might be preferable that the solution of a staining reagent is neutral to avoid pH-dependent deterioration of the samples [9] [Note. Uranyl acetate forms precipitates by neutralization and thus is often used at an acidic pH as an unbuffered solution. Nevertheless, the structure of the materials is, in most cases, well preserved to give adequately stained images, probably because uranyl salts also work as a quick and strong fixative [15], as will be mentioned shortly.] Since acetate salts of lanthanides showed a neutral pH [9] when dissolved as a 1–10% aqueous solution, we evaluated their performance as negative-staining reagents. Figure 3 exhibits the images of Aβ1–42 fibril and T4 phage, negatively stained with samarium acetate, gadolinium acetate and uranyl acetate, respectively. The shape of individual fibril oligomeric particles and the arrangement of various parts of T4 phage were clearly contrasted and appeared like authentic negatively stained images of each material [9]. Upon continuous electron-beam irradiation, no appreciable crystallization was observed within the time we needed for field scanning and taking sufficient number of pictures. All of these favorable properties indicate that new reagents we introduced here might be used as good substitutes for uranyl acetate for negative staining. We also examined whether meglumine gadoterate could be used as another negative stain for macromolecular samples. Unfortunately, however, this inert reagent did not give appreciable image contrast when compared with the other metal reagents (data not shown). Inability of new reagents as a fixative for rotary shadowing of protein polymers Rotary shadowing is another useful technique to obtain high-contrast images of various macromolecules. Target materials are dissolved in glycerol-containing buffer and sprayed onto a flat mica surface [16]. In order to protect unwanted disassembly of the polymerized materials into small pieces by the shear force of spraying, Mabuchi (1991) devised a unique technique utilizing glycerol for material preservation and uranyl acetate as a fixative [12]. Though the method works effectively [13], it requires more uranyl acetate than that is needed in negative staining. We thus examined whether the new metal stains we introduced could substitute uranyl acetate and somewhat help stabilizing the protein assembly. Several trials were made under different conditions, but we did not find any appreciable effects of new heavy metal salts as a fixative (data not shown). Discussion In this report, we proposed two acetate salts of lanthanide, samarium and gadolinium, as new staining reagents for biological electron microscopy and showed that they could be excellent substitutes for uranyl acetate for thin-section staining and a fairly good substitute for negative staining but cannot be substituted as a fixative. The last issue might reflect the fact that these lanthanides are non-hazardous reagents still performing favorably as new staining reagents. Though several compounds have been introduced as substitutes for uranyl acetate in biological electron microscopy, most of them still have some weakness either in the stability of solutions or in the biological safety [5–7,17]. In this sense, our new reagents could almost perfectly replace uranyl acetate, satisfying all the needs for the staining of thin sections. During the course of this study, we observed that the image contrast of the metal-stained sections greatly enhances only when tissues were post-fixed with osmium, strongly suggesting the essential role of osmium pretreatment to the metal-reagent staining. Chemical fixation of tissues and cells by osmium tetroxide has been assumed mainly as hydroxylation or oxidation of the double bond of unsaturated fatty acids and nucleic acids [18]. It is hard to attribute observed enhancement to additional beam scattering effect of bound osmium, because osmium atom by itself has a poor shielding capability. Roth and Hinckley [19] reported that osmium forms various chelating compounds including amino acids of the protein, ions and/or hydrogen bonding to carboxylic and phosphoric acid groups. Since these complexes are insoluble in water or ethanol, they might locally accumulate a large amount of metal reagents to dramatically raise the stain density. In order to evaluate the utility of lanthanide compounds as negative-staining reagents, we examined the performance of their acetate salts as well as that of meglumine gadoterate. Negatively contrasted images obtained by uranyl acetate staining have been thought to be somewhat different from those obtained by other authentic negative-staining reagents, including both negative- and positive-staining effects [9]. It is notable that the last candidate did not show appreciable negative-staining effect, despite the appreciable staining capability for thin sections. The chemical difference from the other compounds is that the metal ion is not exposed but is sitting in the center of the surrounding framework [14]. We thus speculate that strong chelation of the metal ions to the functional groups of the target macromolecules might considerably contribute to the negative-contrast effect exerted by lanthanides, as well as uranyl acetate staining. Though the straight solutions of lanthanide acetates work as right substitutes for uranyl acetate, we noticed the possibility that inclusion of some supplementary substances could improve their staining capability. Further attempts are in progress. Concluding remarks We have introduced two lanthanide salts, samarium and gadolinium triacetate, as right substitutes for environmentally hazardous uranyl stains for biological electron microscopy. They were effective as stains for plastic-embedded animal and plant tissues and as negative stains of macromolecules, but did not work as a fixative. Funding This study was supported partly by the System Development Program for Advanced Measurement and Analysis (Program-S) from Japan Science and Technology Agency from the Ministry of Education, Science, Sports and Culture of Japan to E.K. References 1 Watson M L . Staining of tissue sections for electron microscopy with heavy metals , J. Biophys. Biochem. Cytol. , 1958 , vol. 4 (pg. 475 - 478 ) 10.1083/jcb.4.4.475 Google Scholar Crossref Search ADS PubMed WorldCat Crossref 2 Watson M L . Staining of tissue sections for electron microscopy with heavy metals. II. Application of solutions containing lead and barium , J. Biophys. Biochem. Cytol. , 1958 , vol. 4 (pg. 727 - 730 ) 10.1083/jcb.4.6.727 Google Scholar Crossref Search ADS PubMed WorldCat Crossref 3 Brenner S , Horne R W . A negative staining method for high resolution electron microscopy of viruses , Biochim. Biophys. Acta , 1959 , vol. 34 (pg. 103 - 110 ) 10.1016/0006-3002(59)90237-9 Google Scholar Crossref Search ADS PubMed WorldCat Crossref 4 Bradley D E , Kay D . The fine structure of bacteriophages , J. Gen. Microbiol. , 1960 , vol. 23 (pg. 553 - 563 ) Google Scholar Crossref Search ADS WorldCat 5 Inaga S , Katsumoto T , Tanaka K , Kameie T , Nakane H , Naguro T . Platinum blue as an alternative to uranyl acetate for staining in transmission electron microscopy , Arch. Histol. Cytol. , 2007 , vol. 70 (pg. 43 - 49 ) 10.1679/aohc.70.43 Google Scholar Crossref Search ADS PubMed WorldCat Crossref 6 Sato S , Adachi A , Sasaki Y , Ghazizadeh M . Oolong tea extract as a substitute for uranyl acetate in staining of ultrathin sections , J. Microsc. , 2007 , vol. 229 (pg. 17 - 20 ) 10.1111/j.1365-2818.2007.01881.x Google Scholar Crossref Search ADS WorldCat Crossref 7 Ikeda K-I , Inoue K , Kanematsu S , Horiiuchi Y , Park P . Enhanced effects of nonisotopic hafnium chloride in methanol as a substitute for uranyl acetate in TEM contrast of ultrastructure of fungal and plant cells , Microsc. Res. Tech. , 2011 , vol. 74 (pg. 825 - 830 ) Google Scholar PubMed OpenURL Placeholder Text WorldCat 8 van Bruggen E F J , Wiebenga E H , Gruber M . Negative-staining electron microscopy of proteins at pH values below their isoelectric points: its application to hernocyanin , Biochim. Biophys. Acta , 1960 , vol. 42 (pg. 171 - 172 ) 10.1016/0006-3002(60)90771-X Google Scholar Crossref Search ADS WorldCat Crossref 9 Bradley D E . A study of the negative staining process , J. Gen. Microbiol. , 1962 , vol. 29 (pg. 503 - 516 ) Google Scholar Crossref Search ADS PubMed WorldCat 10 Leeson T S , Higgs G W . Lanthanum as an intracellular stain for microscopy , Histochem. J. , 1982 , vol. 14 (pg. 553 - 560 ) 10.1007/BF01011888 Google Scholar Crossref Search ADS PubMed WorldCat Crossref 11 Reynolds E S . The use of lead citrate at high pH as an electron-opaque stain for electron microscopy , J. Cell Biol. , 1963 , vol. 17 (pg. 208 - 212 ) 10.1083/jcb.17.1.208 Google Scholar Crossref Search ADS PubMed WorldCat Crossref 12 Mabuchi K . Heavy-meromyosin-decorated actin filaments: a simple method to preserve actin filaments for rotary shadowing , J. Struct. Biol. , 1991 , vol. 107 (pg. 22 - 28 ) 10.1016/1047-8477(91)90027-T Google Scholar Crossref Search ADS PubMed WorldCat Crossref 13 Katayama E , Ikebe M . Mode of caldesmon binding to smooth muscle thin filament: possible projection of the amino-terminal of caldesmon from native thin filament , Biophys. J. , 1995 , vol. 68 (pg. 2419 - 28 ) 10.1016/S0006-3495(95)80424-8 Google Scholar Crossref Search ADS PubMed WorldCat Crossref 14 Runge V M , Gelblum D Y , Pacetti M L , Carolan F , Heard G . Gd–HP–DO3 A in clinical MR imaging of the brain , Radiology , 1990 , vol. 177 (pg. 393 - 400 ) Google Scholar Crossref Search ADS PubMed WorldCat 15 Zhao F Q , Craig R . Capturing time-resolved changes in molecular structure by negative staining , J. Struct. Biol. , 2003 , vol. 141 (pg. 43 - 52 ) 10.1016/S1047-8477(02)00546-4 Google Scholar Crossref Search ADS PubMed WorldCat Crossref 16 Tyler J M , Branton D . Rotary shadowing of extended molecules dried from glycerol , J. Ultrastruct. Res. , 1980 , vol. 71 (pg. 95 - 102 ) 10.1016/S0022-5320(80)90098-2 Google Scholar Crossref Search ADS PubMed WorldCat Crossref 17 Yamaguchi K , Suzuki K-I , Tanaka K . Examination of electron stains as a substitute for uranyl acetate for the ultrathin sections of bacterial cells , J. Electron Microsc. , 2010 , vol. 59 (pg. 113 - 118 ) 10.1093/jmicro/dfp045 Google Scholar Crossref Search ADS WorldCat Crossref 18 Hayat M A . , Principles and Techniques of Electron Microscopy , 1970 , vol. Vol. 1 New York an Nostrand Reinhold (pg. 35 - 58 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 19 Roth W J , Hinckley C C . Synthesis and characterization of osmyl-amino acid complexes: molecular structure of trans-dioxobis(glycinato)osmium(VI), OsO2(NH2CH2COO)2 , Inorg. Chem. , 1981 , vol. 20 (pg. 2023 - 2026 ) 10.1021/ic50221a019 Google Scholar Crossref Search ADS WorldCat Crossref © The Author 2011. Published by Oxford University Press [on behalf of Japanese Society of Microscopy]. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
TI - New versatile staining reagents for biological transmission electron microscopy that substitute for uranyl acetate
JF - Journal of Electron Microscopy
DO - 10.1093/jmicro/dfr084
DA - 2011-12-01
UR - https://www.deepdyve.com/lp/oxford-university-press/new-versatile-staining-reagents-for-biological-transmission-electron-AxF8QAk6hx
SP - 401
EP - 407
VL - 60
IS - 6
DP - DeepDyve
ER -