MICROSCOPY RESEARCH AND TECHNIQUE 22:185-193 (1992)

Chemical Extraction of the Cytosol Using Osmium Tetroxide for High Resolution Scanning Electron Microscopy P.J. LEA, M.J. HOLLENBERG, R.J. TEMKIN, AND P.A. KHAN Department of Anatomy, Faculty of Medicine (P.J.L.,M.J.H., R.J.T.),and Department of Zoology (PA.K.1, University of Toronto, Toronto, Ontario, Canada M5S 1A8

KEY WORDS

Cytosol extraction, HRSEM, Fixation, Intracellular organelles

ABSTRACT

Detailed examination of subcellular structures in three dimensions (3D) by high resolution scanning electron microscopy (HRSEM) is now possible due to improvements in the design of the scanning electron microscope and the introduction of methods of specimen preparation using chemical removal of the cytosol and cytoskeleton by dilute osmium tetroxide. Cells which have been fixed, frozen, cleaved, thawed, and subjected to cytosol extraction display intact intracellular structures in 3D including nuclear chromatin, endoplasmic reticulum, mitochondria, and the Golgi complex at a resolution close to that of conventional biological transmission electron microscopy (TEM). Small changes in the 3D structure of subcellular components can be conveniently examined in this way in development, in a variety of physiological processes and in disease. Broad areas of the specimen can be quickly surveyed by HRSEM since sectioning is not required and specimens of comparatively large size (up to 5 mm3) can be placed in the microscope. Extraction of the cytosol with dilute osmium tetroxide (OsOJ exposes subcellular structures in relief, permitting their examination in 3D from several aspects. However, the OsO, extraction technique is limited, since significant intracellular structures, such as the cytoskeleton, vesicles, and antibody binding sites can be removed or inactivated during the cytosol removal steps. 0 1992 Wiley-Liss, Inc.

INTRODUCTION The study of intracellular organelle structure has advanced rapidly in the last few years because of improved methods of specimen preparation. However, until recently, high resolution scanning electron microscopy (HRSEM) has not been the instrument of choice for this work. Instead, such diverse techniques as freeze-fracture with optional freeze-etching (Caldwell et al., 1987), “slam”-freezing (Heuser et al., 1979; Hirokawa, 1986))and conventional transmission electron microscopy (TEM) of both random and serial thin sections combined with computer assisted 3D reconstruction (Lea and Pawlowski, 1986) have been favoured. Computer assisted reconstruction from thin sections and high voltage TEM of thick sections (Frank et al., 1986), as well as the development of high resolution metal coating processes (Peters, 1985, 1986), are techniques which have been exceptionally useful in gaining an appreciation of the 3D microanatomy of the cell’s interior. Together, all of the foregoing techniques, carefully applied, have provided much of our current information and understanding of the microanatomy of the cell in both animal and plant tissues (Barnes and Blackmore, 1986). Since each of these methods offers both advantages and disadvantages to the investigator, it has become clear that the most comprehensive appreciation of cell structure is usually obtained when more than one technique is applied and results compared. Thus, it was in this manner that concepts developed on the ultrastructure of such structures as the cell membrane, the mitochondrion, the Golgi complex and the nucleus.

0 1992 WILEY-LISS, INC.

Yet, it has become clear that these methods by themselves are restricted in their application, since they can only readily survey extremely small areas of the cell, especially in 3D. Therefore, studies of progressive pathological or dynamic processes have remained well beyond the scope of these methods because of the huge amount of work involved. Likewise, it has been difficult to observe structural variations in organelles in cells from different anatomical sites since a great many ultrathin sections would have t o be cut and serially combined. Also, attempts to verify past work have been similarly frustrated. Accordingly, much of what is known of 3D organelle structure is based on few studies in a select number of different tissue types (Bullock, 1984). What has been lacking is a fast comprehensive and reliable method of viewing intracellular components a t both low and high magnification in 3D, at a resolution approximating currently available conventional TEM. Fortunately, there is now a growing body of evidence that HRSEM, when properly applied, can provide just such a method (Hollenberg et al., 1989, Hollenberg and Lea, 1988; 1989, Lea and Hollenberg, 1987, 1988, 1989a,b; Tanaka, 1989). However, the application of this method requires removal of the cytosol during specimen preparation so that intracellular structures can be exposed.

Received January 2, 1991; accepted in revised form April 15, 1991. Address reprint requests to Dr. P.J. Lea, Department of Anatomy, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada M5S 1A8.

Figs. 1 and 2.

HRSEM BY CYTOSOL EXTRACTION

So far, the technique of choice for doing this has been chemical extraction using dilute OsO,, a method pioneered by Tanaka (1989). Our own experience indicates that chemical extraction in this way offers both significant advantages and disadvantages to the investigator. This has led to the conclusion that a variety of new methods of cytosol extraction must now be developed for specific purposes in order to speed the rapid advance in our knowledge of cell structure and function. MATERIALS AND METHODS Tissue Preparation for HRSEM Tissues used were liver from adult Sprague-Dawley albino rats (approximately 300 g), retina from 21-dayold chick embryos, and human kidney. After removal, liver specimens were sliced into strips about 1-2 mm thick and 5 mm long. Some strips were fixed by immersion in 0.5% glutaraldehyde and 0.5% paraformaldehyde in 0.07 M phosphate buffer (pH 7.2) for 15, 30 or 60 minutes at room temperature. Other pieces were fixed for 60 minutes in 1.0% glutaraldehyde and 4.0% paraformaldehyde in phosphate buffer also at room temperature. The chick retina underwent primary fixation by immersion in 0.5% glutaraldehyde and 0.5% paraformaldehyde in 0.07 M phosphate buffer for 75 minutes. The fixative was kept a t a low temperature during this period by immersing the specimen vials in ice. The human kidney was fixed in the same manner as the chick retina except that the time was reduced to 60 minutes. After the primary aldehyde fixation, all tissues were washed two times in phosphate buffer and placed into 1% OsO, in phosphate buffer for 1.5 hours at 20 "C. Following this initial osmium treatment, the specimens were thoroughly washed to remove excess OsO,. The blocks were then placed into 25% dimethylsulfoxide (DMSO) in distilled water for 0.5 hour, followed by another 0.5 hour in 50% DMSO. Meanwhile, Freon 22 was condensed to its liquid form using a liquid nitrogen well. The blocks were frozen by rapid immersion in the liquid Freon 22 and then transferred to liquid nitrogen. A specially machined brass block, immersed in liquid nitrogen, was used to support the cleavage process. The brass block, 100mm in diameter, has a 1 0 m m high rim around its circumference to contain the frozen tissue blocks. The blocks were placed into the brass well and cleaved under liquid nitrogen with a precooled razor blade. The specimen blocks, when originally cut from the relevant tissue, were cut in such a way that

Fig. 1. HRSEM micrograph of a portion of a rat liver hepatocyte. The liver was fixed in 0.5% glutaraldehyde and 0.5% formaldehyde in phosphate buffer for 60 min, freeze-cleaved and extracted in 0.1% osmium tetroxide solution for 1.5 days. The tissue has a dense granular appearance without any recognizable intracellular structures visible, although areas of less dense granularity (MI may correspond to mitochondria. x 39,000. Fig. 2. HRSEM micrograph of another rat liver specimen prepared as in Figure 1, hut fixed for only 15 min with aldehyde. Some membranousstructure is visible but details are still obscured by unextracted protein. x 34,000.

187

following freeze-cleavage, it was still possible to identify the cleaved surfaces with certainty. Thawing of the cleaved blocks took place in vials containing 50% DMSO at 20 "C and the DMSO was then removed by washing in phosphate buffer. The specimens were treated with 1%OsO, in phosphate buffer for 1hour at 20 "C followed by extraction of the cytosol by immersion of the samples in 0.1% OsO,. The liver samples fixed in the lower concentration aldehyde solution were extracted using 0.1% OsO, for 1.5 to 3 days. Those fixed in the higher concentration solution were extracted for 14 days. Both the chick retina and the human kidney were extracted for 3 days. After the extraction period, the specimens were washed in buffer, placed into 1% tannic acid in distilled water for 1hour, washed again, placed into 1% OsO, in buffer for 1 hour and returned to buffer. Dehydration was accomplished through a graded ethanol series. The specimens were dried by a critical point method (Lea and Ramjohn, 1980) and specimens were mounted on aluminum stubs using carbon conductive paint.

Metal Coating of the SEM Samples The specimens were coated in a Polaron sputter coater model E5100 (Polaron Corp. United Kingdom), using argon gas as the ionizing plasma. The thickness of the gold/palladium alloy which was deposited onto the samples was monitored with a modified quartz crystal, thin film measuring device (QSG 201, Balzers High Vacuum, Liechtenstein). The average mass density of the metal film was determined to be equivalent to a thickness of approximately 10 nm (nanometers).It was found that this film thickness gave a n optimal signal to noise ratio for the high resolution imaging mode of the HRSEM. High Resolution Scanning Electron Microscopy Specimens were examined in a Hitachi S-570 scanning electron microscope using a standard tungsten electron source set at a n accelerating voltage of 20 KV and a working distance of 0 to - 4 mm. The negative working distance of - 4 mm means th a t the specimen, when mounted on a 5 mm diameter high resolution specimen stub, was introduced into the bore of the objective lens to a maximum of 4 mm. The upper secondary electron detector mounted in the column above the objective lens was used to collect the secondary electron image. Micrographs were recorded from a high resolution cathode ray tube using type 55 Polaroid film. The images were recorded as high resolution stereo pairs at 125 steradians (a specimen tilt of ten degrees), allowing observation of the tissue, cells and intracellular structures in 3D. The Hitachi S-570 SEM, when used in this way, obtains a resolution of approximately 3 nm, compared to about 2 nm for a transmission electron microscope in routine use, to view thin sections of biological specimens. RESULTS The degree of cytosol extraction with 0.1% osmium tetroxide was found to be dependent on a number of different factors, including the concentration, temper-

188

P.J. LEA ET AL.

Fig. 3. HRSEM micrograph of a portion of a rat liver hepatocyte. The liver was fixed in 0.5% glutaraldehyde and 0.5% paraformaldehyde in phosphate buffer for 30 min, freeze-cleaved and extracted in 0.1% osmium tetroxide for three days. The mitochondria (M) clearly show tubular cristae. The endoplasmic reticulum (R) is obscured by granular protein which covers the remaining intracellular organelles. A portion of the nucleus can be seen (N). The processing was carried out in plastic ware, resulting in plastic contamination (arrows) of the sample. X 38,000.

Fig. 4. Stereo pair HRSEM micrograph prepared as for TEM by fixing in 1%glutaraldehyde, and 4% paraformaldehyde for 1 hour. Following freeze cleavage, the tissue was extracted for 14 days. Mitochondria (MI, rough endoplasmic reticulum (R), and glycogen ( G ) are clearly visible due to the high degree of protein extraction. x 30,000.

Fig. 5. HRSEM stereo pair of a mitochondrion from Figure 4 shows the extraction effects of dilute OsO,. The tubular cristae are reduced in diameter from approximately 30 nm to about 20 nm (arrows). x 51,000.

Fig. 6. HRSEM stereo pair of intracellular organelles observed in a portion of a retinal pigment epithelial cell of a 21 day chick embryo. (P) melanosome, ( G )Golgi. These organelles fixed with 0.5% glutaraldehyde plus 0.5%paraformaldehyde at 4 "C are well preserved. x 32,000.

ature, and length of the primary aldehyde fixation and the duration of extraction. The hepatocytes in Figures 1-3 underwent primary fixation in 0.5% glutaraldehyde and 0.5%paraformaldehyde a t room temperature. Samples fixed for 60 minutes and extracted for 1.5 days had a solid appearance and the various organelles could not be discerned (Fig. 1). Fixation for only 15 minutes as above, with 1.5 days extraction, resulted in some removal of cytosolic protein but structural details were still obscured (Fig. 2). Figure 3 shows a hepatocyte that was fixed for 30 minutes as above and extracted for 3 days. In this sample, enough extraction had occurred to reveal the internal structure of the mitochondria, especially their tubular cristae. The rough endoplasmic reticulum was barely discernible and there was still a great deal of unextracted protein within the cell.

Figures 4 and 5 show hepatocytes that underwent primary fixation in the higher concentration aldehyde solution routinely used for TEM (4%).They were also fixed for 1 hour but the extraction period extended for 14 days, resulting in a high degree of cytosol removal. In these samples, the structure of the mitochondria and rough endoplasmic reticulum was clearly visible (Fig. 4) and stereo pair viewing at higher magnification revealed structural details in 3D (Fig. 5). Figures 6 and 7 show tissues fixed in a weaker aldehyde solution (0.5%) for a t least 1 hour but with the specimen vials immersed in ice. They were extracted for 3 days. This procedure gave good results. The fixation of the membranous structures was good and the cytosol extraction was of sufficient degree to provide adequate details of organelle structure. In Figure 6, the structure of the Golgi apparatus of a chick retinal pig-

190

P.J. LEA ET AL.

Fig. 7. HRSEM micrograph of human kidney cortex taken from a sample obtained with a biopsy punch. The biopsy was fixed in cold 0.5% glutaraldehyde. Intracellular structures are well fixed and easily recognized and compare favourably to their appearance by TEM. (F)glomerular fenestrated endothelium, (B) basement membrane, (V) visceral epithelial cell containing structures such as (M) mitochon-

dria, (G) the Golgi complex, (R) endoplasmic reticulum and (FP) foot processes. x 21,000. Fig. 8. HRSEM stereo pair of human kidney prepared as in Figure 7. x 10,000.

HRSEM BY CYTOSOL EXTRACTION

191

Immersion fixation of tissue blocks, cut to 1 mm3, in 0.5% glutaraldehyde + 0.5% paraformaldehyde, in M/15 PO4 buffer V

1%OsO4, 1.5 hrs, room temp V

Cryoprotect with dimethylsulfoxide (DMSO) 30 min each in 25% and 50% DMSO V

Freeze tissue in melting Freon Freeze cleave in liquid nitrogen V

Thaw in 50% DMSO V

1 % OsO4, 1 hr, room temp V

Extraction: 0.1% OsO4, room temp, 3-6 days V

1 % Tannic acid in distilled water, 1 hr V

1 % Os04 i n M115 PO4 buffer, 1 hr V

Dehydrate in ethanol series v

Critical Point Dry V

Sputter coat with Gold V

SEM Fig. 9. Outline of procedures used by authors for preparation of tissue used for SEM. All steps are carried out a t room temperature unless otherwise indicated. W15 buffer is Sorenson’s Phosphate Buffer at 0.067 M.

ment epithelial cell is revealed. The stacks are plainly visible and the fenestrations of one of the stacks can be seen. Elements of the endoplasmic reticulum and pigment granules can also be seen. The internal structure of a podocyte in a human kidney glomerulus prepared in the same way is shown in Figures 7 and 8. Again, extraction has been excellent and the structure of the Golgi apparatus, rough endoplasmic reticulum, and mitochondria can be seen, as well as their relationships to one another. Note, however, in the above figures, that successful extraction of the cytosol is accompanied by removal of all elements of the cytoskeleton as well. In addition, structures such as vesicles, lying free in the cytoplasm, may also be removed.

DISCUSSION Relatively little is known about the detailed chemical interactions between aldehydes and proteins (Palade, 1952; Johnson, 1985a,b) or OsO, and proteins. Indeed, much of our available information on this topic with OsO, has come from studies involving protein solutions or gels (Emerman and Behrman, 1982) and not from investigations utilizing actual cells. Nielson and Griffith (1979) present evidence that OsO, acts as a fixative or cross-linker on tissue proteins. In addition, they suggest that further protein fixation may be facilitated by cross-linkages of protein-lipid complexes where the OsO, fixed lipid traps, but did not fix associated protein. They indicate that reactions between OsO, and proteins probably occur at the amino or imino N-donor groups of amino acids and a t the sulphur donor sites, or a t the noncyclic, carbon-carbon

double bond in the side chain of tryptophan. MaupinSzamier and Pollard (1978) indicate that the amino acids cysteine and methionine are highly reactive with OsO,, especially a t alkaline pH. This high degree of reactivity may be due to the presence of a double bonded sulphur atom in their side chains. All other amino acids do not contain sulphur, and hence, are not as easily oxidized. These authors stated that tryptophan, histidine, proline, and arginine also react with OsO,, and to a lesser extent, so do lysine, asparagine, and glutamine. They claim that the remaining amino acids, such as alanine and isoleucine, commonly found in proteins, are not reactive. The other amino acids which react with OsO, probably do so because their side chains all have nitrogen atoms. Proline is an exception. The amino group of proline is, however, unique in that it is bonded to two carbon atoms. This may account for its reactivity with OsO,. The nonreactive amino acids simply have carbon (no C = C), hydrogen, and oxygen in their side chains and thus are not oxidized. Evidence in support of this is provided by Maupin-Szamier and Pollard (1978) who note that the sulphur-containing amino acids are oxidized to their sulfone derivatives. Furthermore, it is possible that oxidation of side chains containing nitrogen results in the production of ammonia gas which these authors detected. In addition to protein fixation, the fact that OsO, extracts protein from tissue has long been realized (Lisak et al., 1976; Luft and Wood, 1963). Indeed, Porter and Kallum (1953) used OsO, to remove the cytosol so as t o render membrane bound organelles more suitable for “detailed study.” Bahr (1955) also showed that

192

P.J. LEA ET AL.

exposure to OsO, promoted protein extraction. Moreover, today, dilute and slightly hypotonic solutions of OsO, are routinely used to extract proteinaceous cytosol from cells for HRSEM (Tanaka, 1981; Fukudome and Tanaka, 1986; Tanaka et al., 1986; Lea and Hollenberg, 1989a,b). Further, it is worth noting that protein extraction occurs not only when OsO, is used as a primary fixative, but also when used following prefixation with glutaraldehyde. Indeed, according to Hayat (1981) protein extraction may even be exacerbated when both fixatives are used because of the extra conformational changes imposed by both OsO, and glutaraldehyde on proteins, while such conformational changes may not necessarily occur when each fixative is used on its own. The precise chemistry explaining why prolonged exposure to OsO, extracts protein is not known. It may be by a process of oxidative deamination (Lisak et al., 1976). It is likely that OsO, denatures proteins sufficiently to render them soluble in water and thus, prolonged exposure of tissues to slightly hypotonic solutions of OsO, may cause the cells not only to swell slightly, but also to provide an aqueous medium for these unfixed proteins to diffuse out of the cell. Another effect that OsO, may have is to cleave proteins (see Hayat, 1981). Such cleaved products may not be crosslinked and thus are readily extracted. Because osmium is so useful in preserving ultrastructure, numerous attempts have been made to try to “protect” the cytoskeleton in osmium-fixed tissue. Specifically, Aoki and Tavassoli (1981) reported that the Os0,-thiocarbohydrazide-Os04 method (OTO) minimized, but did not eliminate, actin filament destruction. Another popular method uses Os0,-dimethyl sulfoxide-OsO, (OD01 (Fukudome and Tanaka, 1986). These methods, in addition, have proven to be valuable in increasing conductivity of the sample and, as a result, have the benefit of minimizing or eliminating the need for metal film coating of the sample (Bullock, 1984).The main drawback to all of these methods, however, is that the concentrations of OsO, ( = 1 hour) for cytosol extraction are also sufficient to extract major portions of the cytoskeleton. These procedures are also very time consuming, and their usefulness at high resolution has been questioned (Gabriel, 1982). According to Maupin and Pollard (19831, tannic acid affords protection for a portion of the cytoskeleton (actin) from OsO, either with or without glutaraldehyde prefixation. Murakami and Jones (1980) suggest using the tannic acid-OsO?thiocarbohydrazide-Os0, (TAOTO) method for this purpose (see also Takahashi, 1986). A limitation to these methods is that cells must be pretreated with a nonionic detergent, saponin, (0.5 mg/ml) to permit tannic acid (2 mg/ml) to penetrate the plasma membrane (Maupin and Pollard, 1983). However, using a much higher concentration of tannic acid (80 mg/ml), Maupin and Pollard report that tannic acid by itself can penetrate the cells and moreover, cytosol is extracted. Therefore, it may be worth exposing cells to prolonged doses of concentrated tannic acid, prepared in a stabilization buffer, to test the effectiveness of this acid in extracting cytosol while selectively binding to the cy-

toskeleton. Once the cytosol has been extracted, it may be necessary to further fix the remaining proteins of the cytoskeleton with glutaraldehyde in order to eflectively stabilize them for exposure to conditions involved with freeze-drying and viewing in the HRSEM. If extraction of the cytosol has been efficient, then further glutaraldehyde fixation should not pose a problem in obscuring detail. Our observations of the action of dilute OsO, in the process of cytosol extraction have led to the following hypothesis. We suggest that the oxygen radical (-04) released by the dilute OsO, reacts with the unsaturated lipid sandwiched between the two hydrophobic layers of exposed organelle membranes. The osmium in elemental form is also deposited along the proteinaceous outer layers of the membranes. This layer of osmium probably protects and preserves the structure of the membrane while masking receptor sites. However, protein which is not in continuity with lipid is not protected and is consequently broken down and therefore washed away during the extraction process. This is well illustrated by the fact that the cytoskeleton is extracted when exposed to dilute OsO,. We have found that the efficacy of dilute osmium extraction is dependent on the conditions of the primary aldehyde fixation. Standard fixation with concentrated aldehyde as done for TEM necessitates a very long extraction period that is impractical for routine application of the technique for HRSEM. The use of a more dilute primary aldehyde fixation gives quicker results but the time of fixation must be restricted to between 15 and 30 minutes. A result of this short initial fixation period is that the aldehyde only penetrates a short distance into the tissue block and only the edge of the tissue is adequately preserved. Increasing the period to 60 minutes allows further penetration of the fixative but extraction is slowed due to the greater fixation of the tissue. Use of an ice cold dilute aldehyde solution allows fixation times to be extended to 60 to 90 minutes without the need to increase extraction time, even possibly shortening it. The low temperature slows the action of the aldehyde on the proteins, preventing overfixation. The cold may also break down some of the cytoskeletal elements, such as the microtubules, allowing easier extraction. The longer fixation times may also allow greater penetration of the fixative crucial for good preservation of tissues, such as human biopsies that cannot be fixed by perfusion. In conclusion, despite the superior results obtained by HRSEM methods using dilute osmium to extract cytosol, much remains to be done in order to develop a method to preserve all relevant, intracellular structures and, a t the same time, permit immunocytochemistry inside the cell using HRSEM.

REFERENCES Aoki, M., and Tavassoli, M. (1981) OTO method for preservation of actin filaments in electron microscopy. J. Histochem. Cytochem., 29682-683. Bahr, G.F. (1955) Continued studies about the fixation with osmium tetroxide. Exp. Cell Res., 9:277-285. Barnes, S.H., and Blackmore, S. (1986) Plant ultrastructure in the scanning electron microscope. Scanning Electron Microsc., 1:281289.

HRSEM BY CYTO)SOL EXTRACTION Bullock, B.R. (1984) The current status of fixation for electron microscopy: A review. J . Micros., 133:l-15. Caldwell, R.B., Slapnick, S.M., and McLaughlin, B.J. (1987) Quantitative freeze fracture and filipin binding study of retinal pigment epithelial cell basal membranes in diabetic rats. Exp. Eye Res., 44245-252. Emerman, M., and Behrman, E.J. (1982) Cleavage and cross-linking of proteins with osmium (VIII) reagents. J . Histochem. Cytochem., 30:395-397. Frank, J., McEwen, B., Radermacher, M., Turner, J.N., and Rieder, C.L. (1986) Three dimensional tomographic reconstruction in high voltage electron microscopy. Proc. XIth Int. Cong. Electr. Microsc., Kyoto, pp. 1145-1150. Fukudome, H., and Tanaka, K. (1986) A method for observing structures of free cells by scanning electron microscopy. J. Micros., 141: 171-178. Gabriel, B.L. (1982) Biological Scanning Electron Microscopy. Van Nostrand Reinhold, Toronto. Hayat, M.A. (1981)Fixation for Electron Microscopy. Academic Press, Toronto. Heuser, J.E., and Salpeter, S.R. (1979) Organization of acetylcholine receptors in quick-frozen, deep-etched, and rotary-replicated Torpedo postsynaptic membrane. J . Cell Biol., 82150-173. Hirokawa, N. (1986) Quick-freeze deep-etch replica observations of the cytoskeleton. Proc. XIth Int. Cong. Electr. Microsc., Kyoto, pp. 1901-1904. Hollenberg, M.J., Cormack, D.H., and Lea, P.J. (1989) Stereo Atlas of the Cell. B.C. Decker, Toronto, Philadelphia. Hollenberg, M.J., and Lea, P.J. (1988) High resolution scanning electron microscopy of the retinal pigment epithelium and Bruch's layer. Invest. Ophthalmol. Vis. Sci., 29:1380-1389. Hollenberg, M.J., and Lea, P.J. (1989) Advantages of visualization of organelle structure by high resolution scanning electron microscopy. Scanning, 11:157-168. Johnson, T.J.A. (1985a) Glutaraldehyde fixation chemistry. A scheme for rapid crosslinking and evidence for rapid oxygen consumption. In: Science of Biological Specimen Preparation. M. Mueller, R.P. Becker, A. Boyde, and J.J. Wolosewick, eds. Sem Inc., Chicago, U.S.A., pp. 51-62. Johnson, T.J.A. (198513) Aldehyde fixatives: Quantification of acid producing reactions. J . Electron Microsc. Tech., 2:129-138. Lea, P.J., and Hollenberg, M.J. (1987) High resolution 3-dimensional scanning electron microscopy of rat hepatocytes. Proc. Microsc. SOC. Canada, 1424-25. Lea, P.J., and Hollenberg, M.J. (1988) High resolution scanning electron microscopy (HRSEM) of mitochondria from various rat tissues. Proc. Electr. Microsc. SOC.Am., 4614-15. Lea, P.J., and Hollenberg, M.J. (1989a) Mitotochondrial structure revealed by scanning electron microscopy (SEM). In: Cells and Tissues: A Three Dimensional Approach by Modern Techniques in Microscopy. P.M. Motta, ed. Alan R. Liss, New York, pp. 63-70. Lea, P.J., and Hollenberg, M.J. (1989b) Mitochondria1 structure re-

193

vealed by high-resolution scanning electron microscopy. Am. J. Anat., 184:245-257. Lea, P.J., and Pawlowski, A. (1986) Human melanocytic nevi. I. Electron microscopy and three dimensional computer reconstruction of nevi and basement membrane zone from ultrathin serial sections. Acta Derm. Venereol., 127:5-15. Lea, P.J. and Ramjohn, S. (1980) Investigating the substitution of ethanol with liquid carbon dioxide during critical point drying. Microsc. Acta, 83991-295. Lisak, J.C., Kaufman, H.W., Maupin-Szamier, P., and Pollard, T.D. (1976) The action of OsO, on proteins and amino acids. Biol. Bull. (Woods Hole), 151:418. Luft, J.H., and Wood, R.L. (1963) The extraction of tissue protein during and after fixation with osmium tetroxide in various buffer systems. J . Cell Biol., 19:46A. Maupin, P., and Pollard, T.D. (1983) Improved preservation and staining of HeLa cell actin filaments. Clathrin coated membranes and other cytoplasmic structures by tannic acid-glutaraldehyde-saponin fixation. J . Cell Biol., 96:51-62. Maupin-Szamier, P., and Pollard, T.D. (1978) Actin filament destruction by osmium tetroxide. J . Cell Biol., 77:837-852. Murakami, T., and Jones, A.C. (1980) Conductive staining of biological specimens for non-coated scanning electron microscopy: Double staining by tannin-osmium-thiocarbohydrazide-osmiummethods. Scanning Electron Microsc., 1980, I:221. Nielson, A.J., and Griffith, W.P. (1979) Tissue fixation by osmium tetroxide. A possible role for proteins. J. Histochem. Cytochem., 27:997-999. Palade, G. (1952) A study of fixation for electron microscopy. J . Exp. Med.., 95:285-298. - - --- ---

Peters, K.-R. (1985) Working at higher magnification in scanning electron microscopy with secondary and backscattered electrons on metal coated biological specimens and imaging macromolecular cell membrane structures. Scanning Electron Micros., 4:1519-1544. Peters, K.-R. (1986) Rationale for the application of thin, continuous metal films in high magnification electron microscopy. J. Microsc., 14225-34. Porter, K.R., and Kallum, F. (1953)The properties and effects of osmium tetroxide as a tissue fixative with special reference to its use for electron microscopy. Exp. Cell Res., 4:127-141. Takahashi, G. (1986) Tannin-ferrocyanide-OsO, method for transmission electron microscopy. Proc. XIth Int. Cong. Elect. Micros., Kyoto, pp. 2175-2176. Tanaka, K. (1981) Demonstration of intracellular structure by high resolution scanning electron microscopy. Scanning Electron Microsc., 1981, 1I:l-8.Tanaka. K.. Mitsushima. A.. Fukudome. H.. and Kashima. Y. (1986) Three dimensional architecture of the Gdgi complex observed by high resolution scanning electron microscopy. J . Submicros. Cytol., 18:1-9. Tanaka, K. (1989) High resolution scanning electron microscopy of the cell. Biol Cell, 65:89-98.

Chemical extraction of the cytosol using osmium tetroxide for high resolution scanning electron microscopy.

Detailed examination of subcellular structures in three dimensions (3D) by high resolution scanning electron microscopy (HRSEM) is now possible due to...
4MB Sizes 0 Downloads 0 Views