Transmission electron microscopy of critical point dried tissue after observation in the scanning electron microscope.код для вставкиСкачать
BRIEF COMMUNICATION Transmission Electron Microscopy of Critical Point Dried Tissue after Observation in the Scanning Electron Microscope ' SAMUEL M. MELLER, MICHAEL R. COPPE, SUSUMU IT0 AND ROBERT E. WATERMAN Department of Anatomy, Harvard Medical School and Harvard School of Dental Medicine, Boston, Massachusetts 021 15 and Department of Anatomy, University of New Mexico, School of Medicine, Albuquerque, New Mexico 87131 Embryonic and adult rodent tissues were fixed and prepared for scanning electron microscopy by dehydration in ethanol followed by critical point drying with liquid carbon dioxide or Freon 13 (E. I. du Pont de Nemours, Inc. 1. After coating the dried specimens with evaporated metal, the tissues were studied by scanning microscopy. The same tissues were subsequently embedded in Epon-Araldite, thin sectioned and examined by transmission electron microscopy. The cytological details in these specimens were comparable to tissues embedded directly, without drying or metal-coating. With this technique it is possible to identify with greater certainty the structures responsible for surface contours revealed by the scanning electron microscope. ABSTRACT Much useful information can be obtained by examining the surface of a biological specimen with the scanning electron microscope (SEM). However, there is often some uncertainty a s to whether a particular surface contour is actually part of the underlying cell or adherent extraneous material. Moreover, with scanning microscopy the precise identity of underlying components that are responsible for surface contours is purely conjectural. Thus there are obvious advantages in being able to examine the same specimen with the transmission electron microscope in thin sections cut perpendicular to the surface previously studied by scanning electron microscopy. This allows definitive identification of the structures responsible for the surface contours. There have been several reports of preparations for scanning electron microscopy that were subsequently viewed in thin section by transmission electron microscopy. However, there was considerable distortion and loss of ultrastructural detail due to air drying (Barber and Boyde, '68) or freezedrying (Wickham and Worthen, '73). Partial infiltration of the tissue with plastic prior to metal coating helps to preserve ANAT. REC.,176: 245-252. cellular details (Cleveland and Schneider, '69) but this method has its limitations with regard to observation of the surface details. The use of critical point drying with liquid C02 (Anderson, '56) or Freon (Cohen, Marlow and Garner, '68) is now widely accepted as the method of choice for drying biological specimens. This method avoids distortion by the high surface tension forces of air drying and ice damage in freeze drying. In a recent paper Wickham and Worthen ('73) compared thin sections of material dried by the critical point method using Freon 13 with sections of specimens prepared by freezedrying. They reported good ultrastructural preservation of cells only after critical point drying. They did not, however, illustrate the detailed preservation of membrane structure in their critical point dried tissue or demonstrate the metal coating on the specimen in sections. In the present paper we reaffirm the superiority of the critical point drying method for SEM preparation and illustrate the retention of ultrastructural details in specimens viewed Received April 6 '73. Acce ted May 11, '73. 1 This work was bupported [y Ford Foundation grant 710-0036, NIDR grant T01-DE00278,and NIH grant AM 7578. 245 246 MELLER, COPPE, I T 0 AND WATERMAN over an hour period, infiltrated with EponAraldite 6005 (Mollenhauer, '64) and propylene oxide ( 1:1) for three hours, and embedded in Epon-Araldite which was then MATERIALS AND METHODS cured at 60°C for 36 hours. Thin sections Golden hamster embryos (LVK strain, were prepared on a Sorvall MT-1 microCharles River Breeding Laboratories, Inc.) tome, stained with uranyl acetate and lead were removed from the uterus 8.5 days citrate and examined in JEOL 100 B or post-coitally and placed in 3% glutaralde- AEI 6 B transmission electron microscopes hyde in 0.1 M phosphate buffer (pH 7.4) (TEM). Control samples of both hamster periat room temperature for three to four hours. After a rinse in the same phosphate derm and mouse jejunum were processed buffer, the tissues were postfixed in 1% through dehydration as described above osmium tetroxide in 0.1 M phosphate buf- and not prepared €or critical point drying fer at room temperature for one to two nor metal coating. Subsequently the tishours and dehydrated at room temperature sues were placed in propylene oxide at through graded ethanol concentrations (35, room temperature for 15 minutes, then 50, 70, 95, 100% ). After complete dehy- propylene oxide : Epon-Araldite ( 1 : 1) for dration, infiltration with iso-amyl acetate four to six hours, and finally embedded i n for two hours was followed by critical point Epon-Araldite and cured at 60°C for 36 drying with liquid CO, in a critical point hours. Appropriate thin sections were examined. bomb (Anderson, '56). Adult Swiss-Webster mouse (Charles RESULTS AND DISCUSSION River Breeding Laboratories, Inc.) jejunum Scanning electron microscopy of liquid was placed in a 0.1 M cacodylate buffered (pH 7.4) 2% paraformaldehyde 3% glu- CO, critical point dried embryos revealed taraldehyde fixative containing 0.01% tri- numerous microvilli which project from nitrocresol (It0 and Karnovsky, '68) a t the surface of the periderm cells. Along room temperature for two hours, and post- the borders of adjacent cells, many microfixed with cacodylate buffered 1% osmium villi were arranged in rows that sharply tetroxide for two hours. The tissue was demarcated rhe junctional areas between washed in 0.05 M maleate buffer (pH 5.4) cells (figs. 1, 2 ) . Thin sections of such and treated with 1% uranyl acetate in areas studied by TEM clearly revealed 0.05 M maleate buffer (pH 6.0) for one these microvilli. However, in thin sections hour and dehydrated with ethanol. After their profiles cannot easily be distinguished dehydration the tissue was infiltrated with from surface folds or plications. Fixed increasing concentrations of Freon TF adult mouse intestine, dried by the critical (E. I. duPont de Nemours, Inc. also known point method using Freon 13, was exas Freon 60) in ethanol and finally in amined in the SEM. On the luminal sur100% Freon TF for one half hour. face of the absorptive cells numerous The specimens were then dried by the hemispherical protrusions which correcritical point method using Freon 13 in sponded to the tips of the microvilli forma Bomar critical point apparatus (SPC ing the striated border were seen (fig. 5). 900) and were affixed to aluminum slugs Confirmation of these structures as the with silver conductive paint and placed on apices of microvilli was made by thin secan Omnirotating stage in a Denton tioning the same material. General cytoVacuum DV 504 evaporator. When a mini- logical preservation was retained and at torr was at- high magnification the trilaminar appearmum vacuum of 5 X tained, a coating of gold or gold-palladium ance of membranes is apparent (figs. 6, (60:40) was evaporated on the rotating 7). Nuclear and cytoplasmic ultrastructure specimens. Coated specimens were exam- (figs. 3, 7 ) did not seem to be significantly ined in a JEOL JSM U-3 SEM at 25 KV u p altered when compared to the same tisto two hours. After observations were made sue prepared by conventional methods in the SEM, the specimens were placed for transmission electron microscopy into propylene oxide, changed 3 or 4 times (figs. 8, 9). i n the SEM that are comparable to tissues processed in the usual manner and not dried prior to embedding. TEM OF CRITICAL POINT DRIED TISSUE The gold or gold-palladium coating (figs. 4, 6 ) is a 250-500 A thick layer of heterogeneously sized particles. In the intestine, only the rounded tips of the microvilli and the glycocalyx between microvilli are coated with gold-palladium. This explains; the hemispherical appearance of these microvilli in the SEM. The metal coating on the embryo was not found beneath densily packed surface projections (fig. 2 ) nor on the under surface of long horizontally oriented microvilli. Despite this uneven coating, no tissue “charging,” indicative of tissue damage due to beam absorption, was noted in this specimen. Positive and direct correlation of SEM and TEM observations on the same specimen allows study of the surface in three dimensions with subsequent confirmation of ultrastructural cytology. If the underlying structure in a particular SEM field required further identification, it is possible to mark an area with a micromanipulator so that this precise area can be sectioned for study by transmission electron microscopy. The general preservation in critical point dried tissue examined by scanning electron microscopy and subsequently embedded and sectioned for transmission electron microscopy is indistinguishable from similar tissues prepared for conventional transmission electron microscopy. Both the carbon dioxide and Freon critical point drying are practical, routine procedures. ’Tissue may be stored in iso-amyl acetate for several weeks without apparent loss of cytological integrity, however, prolonged storage in Freon TF seems to result in a loss of the trilaminar appearance of membranes. Although the reason for this altered appearance is not clear, i t may be due to extraction of membrane components by Freon TF. If prolonged Freon TF immersion is avoided, CO, and Freon critical point drying results in comparable retention of cytoplasmic ultrastructure. If appropriately fixed soft tissues are dried by the critical point method for scanning electron microscopic study and then embedded for transmission electron microscopic study, some of the uncertainties of misinterpreting surface contours seen in 247 the SEM could be avoided. Furthermore, for rare specimens, meaningful information may be obtained from the same specimen by both methods without sacrificing loss of fine detail. An alternative method for using both types of microscopical observations on the same specimen was recently reported by Erlandsen, Thomas and Wendelschafer (’73) who remove the plastic with sodium methoxylate from tissue samples embedded for thin section transmission electron microscopy, and coat them with metal after critical point drying. The same sample may at a later time be reembedded for thin sectioning. It is clear that if tissues are adequately fixed and appropriately prepared, the same sample may be used for both SEM and TEM study. The possibility of being able to modulate between these different types of microscopy of the same specimen with little loss of ultrastructure adds greater credibility to the interpretation of observations by either method alone. LITERATURE CITED Anderson, T. F. 1956 Electron Microscopy of Microorganisms. In: Physical Techniques in Biological Research. Vol. 111. G. Oster and A. Pollister, eds. Academic Press, New York, pp. 177-240. Barber, V. C., and A. Boyde 1968 Scanning electron microscopic studies of cilia. Z. Zellforsch. Mikrosk. Anat., 84: 269-284. Cleveland, P. H., and C. W. Schneider 1969 A simple method of preserving ocular tissue for scanning electron microscopy. Vision Res., 9: 1401-1402. Cohen, A. L., D. P. Marlow and G. E. Garner 1968 A rapid critical point method using fluorocarbons (“Freons”) as intermediate and transitional fluids. J. Microscopie, 7: 331-342. Erlandsen, S. L., A. Thomas and G. Wendelschafer 1973 A simple technique for correlating SEM with TEM on biological tissue originally embedded i n epoxy resin for TEM. Scanning Electron Microscopy/l973 (Part 111) Scanning Electron Microscopy in Pathology, IIT Research Institute, Chicago, pp. 349-356. Ito, S., and M. J. Karnovsky 1968 Formaldehyde-glutaraldehyde fixatives containing trinitro compounds. J. Cell Biol., 39: 168a-169a. Mollenhauer, H. H. 1964 Plastic embedding mixtures for use i n electron microscopy. Stain Techn., 39: 111-114. Wickham, M. G., and D. M. Worthen 1973 Correlation of scanning and transmission electron microscopy on the same tissue sample. Stain Teehn., 48: 63-68. PLATE 1 EXPLANATION OF FIGURES Hamster embryonic periderm critical point dried with liquid COZ. 248 1 A representative area of the head periderm of a n 8.5 day hamster embryo reveals rows of microvilli prominent along cell borders. x 3,200.Inset. Microvilli at higher magnification x 13,000. 2 Taansmission electron micrograph of a portion of a periderm cell with a junctional area showing microvilli coated with gold (arrow) x 17,000. 3 Transmission electron micrograph of nucleus with nucleolus, nuclear envelope and pore exhibiting representative ultrastructural integrity. x 32,000. 4 Transmission electron micrograph of surface of periderm cell with particulate gold coating measuring 250-500 A in thickness and subjacent Golgi cisternae. x 108,000 TEM OF CRITICAL POINT DRIED TISSUE Meller, Coppe, Ito and Waterman PLATE 1 249 PLATE 2 EXPLANATION OF FIGURES Adult mouse jejunal absorptive cells critical point dried with Freon 13. 250 5 Scanning electron micrograph of the brush border of two adjacent intestinal epithelial cells. x 22,000. 6 Transmission electron micrograph of microvilli with gold-palladium coating 250-500 A thick on their termha1 contours, A thin coating of gold-palladium between microvilli (arrow) is deposited on the glycocalyx. Unit membrane preservation is seen. x 70,000. 7 Transmission electron micrograph of absorptive cell exhibiting goldpalladium coating and good cytological preservation. x 12,000. TEM OF CRITICAL POINT DRIED TISSUE Meller, Coppe, Ito and Waterman PLATE 2 25 1 PLATE 3 TEM OF CRITICAL POINT DRIED TISSlJE Meller, Coppe, Ito and Waterman EXPLANATION OF FIGURES Control tissues. 8 Hamster embryonic periderm, fixed and dehydrated like the hamster tissue for the SEM, not prepared for critical point drying or metal coating, and embedded in Epon-Araldite. An area of junctional microvilli is illustrated. x 21,000. 9 Mouse jejunum processed for the TEM directly illustrating ultrastructural preservation comparable to that seen in figure 7. Swollen Golgi cisternae are characteristic of mouse jejunal absorptive cells with the fixation described. 252 x 15,000.