STUDIES on SEEDS I. Fixation of Seeds

STUDIES ON SEEDS

I . Fixation of Seeds

HILTON H . MOLLENHAUER and CLARA TOTTEN

From the Charles F. Kettering Research Laboratory, Yellow Springs, Ohio 45387 Downloaded from http://rupress.org/jcb/article-pdf/48/2/387/1069571/387.pdf by guest on 28 September 2021

ABSTRACT Several fixation procedures were studied to determine those most suitable for preservation of seeds during late stages of development and early stages of germination . These are the periods when the tissues are partially dehydrated and are most difficult to fix for electron microscopy . It was found that a prefixation with a mixture of glutaraldehyde, recon- stituted formaldehyde (i.e. paraformaldehyde), and acrolein, followed by a postfixation in Os04 or KMnO4, gives very acceptable images . The results also indicate that glutaraldehyde is necessary for preservation of cell shape, paraformaldehyde for stabilization of reserve proteins, and acrolein for rapid penetration of tissues . Phosphate, cacodylate, and collidine are all acceptable buffers, although collidine gives the most consistent results .

INTRODUCTION

Processes for the breakdown, transport, and germination . In this report the figures are taken from utilization of reserve protein and lipid, and for the pea (var . Alaska), bean (var. Topcrop), and maize generation of membranes are initiated even before (var. Earliking 108) . Because pretreatment of seeds the seed is fully imbibed . These changes in bio- with antifungal agents or growth substances alters logical activity are accompanied by structural cellular form during early stages of germination, only untreated seeds were used changes, many of which can be resolved by elec- . Seeds were germinated on moist filter paper, in the tron microscopy if the fixation image is sufficiently dark, at 27 °C, for the first 24 hr. For times longer than good. Unfortunately, the fixation processes avail- 24 hr, the seeds were transferred to moist vermiculite, able to the electron microscopist are limited in in the dark, at 27°C. number and many are not adequate for the preser- Tissues were prefixed for 1-2 hr at room temper- vation of seed tissues . The primary problem seems ature (unless otherwise specified in the figure caption) to be related to the rate of fixative penetration . in glutaraldehyde, paraformaldehyde, or acrolein, or The density of cell walls and secreted slimes and combinations of these fixatives . The fixatives were waxes, as well as the relatively dehydrated state of buffered to pH 6 .8-7 .4 by cacodylate, phosphate or seed tissues, act as barriers to the penetration of collidine buffer to which 0.06 M sucrose was some- . Following prefixation, the tissues were fixatives and plastics . The result is that seed tissues times added rinsed in five 8 min changes of buffer at room temper- are difficult to process for electron ature and then postfixed in either 1 .5% Os04 at microscopy . room temperature for 2 hr, or in 1-2% KMn04 at ice-bath temperature for 0.25-0.5 hr. The tissues MATERIALS AND METHODSMETH were then rinsed in several changes of distilled water A large variety of seed tissues have been examined and dehydrated for 7-8 min each in 25, 50, 75 and during late stages of development and early stages of 95% ethanol, and in three 7-8 min changes of 100%

THE JOURNAL OF CELL BIOLOGY . VOLUME 48, 1971 . pages 357-391 387 Downloaded from http://rupress.org/jcb/article-pdf/48/2/387/1069571/387.pdf by guest on 28 September 2021

FIGURE 1 Pea seed soaked 4 .5 hr . The cotyledon was prefixed in 2 .5 0%o glutaraldehyde, postfixed in 1 .5% Os04, and buffered in collidine with 0 .06 M sucrose. Glutaraldehyde is an excellent general fixative for maintenance of cell and membrane form but does not stabilize some of the protein body matrix material . As a result, the unstabilized portions of the matrix material are lost or are displaced within the protein bodies by aggregation or clumping (compare Figs . 1 and 2) . Lipid vesicle (L) ; mitochondrion (M) ; protein body (PB) ; wall (W). X 20,000 .

FIGURE 2 Pea seed soaked 4 .5 hr. The cotyledon was prefixed in 2 .5% paraformaldehyde-2.5%o acrolein, postfixed in 1 .5°10 Os04, and buffered in collidine with 0.06 M sucrose. Paraformaldehyde does not ade- quately stabilize the cell cytoplasm (or cytoskeleton), and most cells shrink during fixation and embedding. Preservation of membranes and organelle form, however, is relatively good though image contrast is usually low . Protein bodies are well preserved and the reserve proteins (i.e. the matrix materials) are not noticeably extracted. The same general image is obtained when only parafornaldehyde is used as the Axative . Lipid (L); plastid (P) ; protein body (PB) ; wall (W) . X 19,000 . Downloaded from http://rupress.org/jcb/article-pdf/48/2/387/1069571/387.pdf by guest on 28 September 2021

FIGURE 3 Pea seed in late development. The cotyledon was prefixed in 3% glutaraldehyde-1 .5% paraformaldehyde-1.5% acrolein, postfixed in 1 .5% 0804, and buffered in 0 .05 M cacodylate . The buffer, as well as the fixative, can materially affect tissue preservation . With cacodylate, the vacuoles and protein bodies often have a dense, nonuniform matrix as shown here (compare Figs . 3 and 4) . X 16,000.

FIGURE 4 Pea seed in late development . The cotyledon was fixed as described in Fig. 3, except that 0 .05 M collidine was used as buffer. The most striking difference is related to the protein bodies which now have a uniform distribution of matrix material (compare with Fig. 3) . It is thought that collidine produces the more reliable image since the image is similar to that following KMnO4 fixation . The other cellular components do not differ markedly as a result of using either cacodylate or collidine as a buffer . X 13,000. Downloaded from http://rupress.org/jcb/article-pdf/48/2/387/1069571/387.pdf by guest on 28 September 2021

FIGURE 5 Maize seed soaked 48 hr. The embryo (in the region of the root cap) was prefixed in 2Y2 % glutaraldehyde and postfixed in 1% 0804, both fixatives being buffered in 0 .05 M phosphate with 0 .5% glucose and maintained at a temperature of 0 °-2°C. Most of the protein body matrix material is extracted by the fixative solutions, and the protein bodies (PB) appear as vacuoles. Glutaraldehyde is not an effec- tive fixative for some protein bodies, especially if the fixative solution is kept at ice bath temperature . X 14,000. FIGURE 6 Pea seed soaked 48 hr . The cotyledon was prefixed in 3°ßo glutaraldehyde-3% acrolein and postfixed in 1°/0 0804, both fixatives being buffered in 0 .05 M collidine with 0 .06 M sucrose. The swelling (see arrows) of some portions of the endoplasmic reticulum is common following glutaraldehyde and Os04 fixation . It does not occur with KMn04 fixation and is therefore thought to be an artifact . Protein body (PB) ; cell wall (W). X 23,000. Downloaded from http://rupress.org/jcb/article-pdf/48/2/387/1069571/387.pdf by guest on 28 September 2021

FIGURE 7 Bean seed in late development . The cotyledon was prefixed in 3% glutaraldehyde-1 .5% paraformaldehyde-1 .5% acrolein (in collidine buffer) and postfixed in aqueous 2% KMn04 . If the fixation time in KMn04 is kept short, then part of the ribosomes and dictyosome-derived coated vesicles are retained . Extraction of ribosomes is preferential, in that those associated with the endoplasmic reticulum are retained longer than those free in the cytoplasm. This is probably due to the more exposed state of free ribosomes to KMnO4, though a protective effect from materials which bond ribosomes to membranes must also be considered. The disposition and form of cellular components are the same as those following 0804 postfixation, though the over-all image is basically that following KMnO4 . Microbody (MB) ; plastid (P) ; protein body (PB). X 12,500. acetone. (Note : Acetone can be used in place of (DMP-30) in a graduated test tube, and shaking ethanol if desired.) vigorously . A fresh plastic mixture was made for The tissues were embedded in a mixture of Epon each step (i.e. three times) in the embedding pro- and Araldite epoxy resins as described previously (11) . cedure. (Note : Other plastic formulations [4, 7, This combination of epoxy resins gives good tissue 19] can be used equally well.) preservation and is exceptionally easy to trim and The dehydrated tissues were soaked for 1-2 hr at section. A complete plastic mixture was made by room temperature in a 50% plastic-50% acetone combining 2 ml of No. 1 plastic stock (consisting of mixture, and then for 2-18 hr in a 75% plastic-25% 62 ml of Epon 812, 50 ml of Araldite 502, and 8 nil of acetone mixture. The tissues were then transferred Aroclor 1248 [Monsanto Co., Organic Chemical Div ., to 100% plastic in flat aluminum foil boats, placed in St. Louis, Mo.] or dibutyl phthalate) with 2 .5 ml of a vacuum oven at 70-80 °C, and evacuated until dodecenylsuccinic anhydride (DDSA) and 4-6 drops bubbles ceased to form (about 5-8 min) . The boats (approximately) of dimethylaminomethyl-phenol (with plastic and tissues) were then transferred to a

HILTON H . MOLLENHAUER AND CLARA TOTTEN Studies on Seeds. I 391 Downloaded from http://rupress.org/jcb/article-pdf/48/2/387/1069571/387.pdf by guest on 28 September 2021

FIGURE 8 Pea seed soaked 2 hr. The cotyledon was prefixed in 3% glutaraldehyde-317 paraformaldehyde-1 .5% acrolein (in collidine buffer with 0 .06 M sucrose) and postfixed in 1% KMn04 (in cacodylate buffer) . Potassium permanganate is an excellent preservative of protein bodies, vacuoles, and membranes . Moreover, it gives an accurate image which can be rapidly scanned and assimilated . Lipid (L) ; protein body (PB) ; starch (ST) ; wall (W) . X 23,500 . FIGURE 9 Bean seed in late development . The cotyledon was prefixed in 3% glutaraldehyde-1 .5% paraformaldehyde-1 .5 170 acrolein (in collidine buffer) and postfixed in 1 170 KMn04 (in cacodylate buffer) . With short time KMn04 fixation, portions of ribosomes and dictyosome-derived coated vesicles (see arrows) are retained. The over-all form of the cell constituents is the same as that following Os04 fixation . Dictyosome (D) . X 33,500 . 60°C oven at atmospheric pressure and left until the Phosphate, cacodylate, and collidine are all plastic hardened (about 18-24 hr) . acceptable buffers for ultrastructural work, though Sections were cut with a Porter-Blum Model MT-1 collidine gives the most consistent results with seed microtome and viewed with a Philips EM-200 elec- tissues . This may be due to the fact that collidine tron microscope . extracts some cellular constituents (8), which would then allow the fixatives and the plastics to RESULTS AND DISCUSSION penetrate more readily . Both fixative and buffer are important in deter- A number of embedding tests were run in con- mining the quality of cellular preservation, and a junction with the fixation studies . The results great divergence of effects is obtained by varying indicate that any of the standard epoxy formula- the fixation procedure, type of tissue, or stage of tions (see references 4, 7, 11, and 19 for examples) cellular development . Some of these effects are work well and are superior to the more fluid illustrated in Figs . 1-9, and discussed briefly in the methacrylate mixtures (for example, see reference accompanying figure captions. 9). Penetration of epoxy resins does not seem to be Fixation of tissues for electron microscopy is a problem, especially when a brief vacuum im- almost totally restricted to reactions with Os04 pregnating cycle is used when the tissues are in the Downloaded from http://rupress.org/jcb/article-pdf/48/2/387/1069571/387.pdf by guest on 28 September 2021 (2, 3, 17, 18, 21-23), aldehydes (6, 14, 16, 17, 22) . 100% plastic mixture (11) . and permanganates (2, 5, 10, 18, 22, 25) . There- It should be emphasized that no single fixative, fore, no other chemical procedures were surveyed or combination of fixatives, is suitable for all in this study . The freeze-etch techniques (1, 12, 13, tissues, and it is always necessary to optimize the 20, 24), use of inert dehydrating agents (15), and fixation procedure for each type of tissue and/or scanning microscopy (Yatsu, personal communica- each stage of cellular development . When this is tion) were considered as possible alternatives to the done, the resulting cellular preservation is sur- chemical preservation of seeds . Freeze-etch tech- prisingly good and quite sufficient for many studies niques, particularly, should be admirably suited of developing and germinating seeds . to the study of resting seeds because the low water content of seeds should alleviate any requirement Charles F . Kettering Contribution No . 392 . Sup- ported in part by United States Public Health Service for pretreatment in glycerol or other solutions . Grant GM15492 . Preliminary results, however, were not satisfactory, Received for publication 4 April 1970, and in revised form and a number of technical problems still await 29 September 1970. resolution . KMn04 and Os04 are not very good fixatives for seeds if used alone, i.e., if used as single-step REFERENCES fixatives without an aldehyde prefixation . KMn04 penetrates poorly, with the result that fixation is 1 . BRANTON, D. 1966 . Fracture faces of frozen mem- inadequate and the electron microscope image branes. Proc. Nat . Acad. Sci . U.S.A. 55:1048 . . Studies on the reactions of OSO4 is granular and often distorted . Os04 penetrates 2. HAKE, T . 1965 and KMn04 with amino acids, peptides and better than KMn04 and often produces a usable proteins. Lab . Invest. 14:470. ultrastructural image . However, the results are 3. KORN, E. D . 1966 . Structure of biological mem- still not as good, nor as consistent, as they are when branes. Science (Washington) . 153 :1491 . two-part or two-step fixatives are used . 4. KUSHIDA, H. 1966 . New embedding with D .E.R. The best preservation of seed tissues is obtained 736 and Epon 812. Proc. 6th Int. Congr . Electron when an aldehyde, or a combination of aldehydes, Microsc. 33. is used as a prefixative and KMn04 or Os04 is 5. LUFT, J. H. 1956 . Permanganate-a new fixative used as a postfixative . One of the most useful pre- for electron microscopy . J. Biophys. Biochem . fixatives is a combination of acrolein, paraformal- Cytol. 2:799. 6. LUFT, J. H . 1959 . The use of acrolein as a fixative dehyde, and glutaraldehyde. The acrolein allows for light and electron microscopy. Anat. Rec. fast and deep penetration (6), the paraformalde- 133:305. hyde preserves protein bodies (see Fig. 2), and 7. LUFT, J . H. 1961 . Improvements in epoxy resin glutaraldehyde maintains over-all cellular form embedding methods. J. Biophys. Biochem . Cytol. (see Fig. 1) . In many instances, acrolein may not 9 :409 . be necessary, and should be deleted from the 8. LUFT, J. H., and R. L . WOOD . 1963 . The extrac- mixture if possible . tion of tissue protein during and after fixation

HILTON H. MOLLENHAUER AND CLARA TOTTEN Studies on Seeds . I 393

with osmium tetroxide in various buffer sys- enzyme histochemical studies with the electron tems . J. Cell Biol. 19(2) :46 A. microscope . J. Histochem. Cytochem . 12 :57 . 9. MOHR, W. P ., and E. C . COCKING . 1968. A 18. SCHIDLOVSKY, G . 1965 . Contrast in multilayer method of preparing highly vacuolated, systems after various fixations. Lab. Invest. 14 : senescent, or damaged tissue for ultrastruc- 475 . tural study. J. Ultrastruct . Res. 21 :171 . 19 . SPURR, A. R. 1969. A low-viscosity epoxy resin 10 . MOLLENHAUER, H . H . 1959. Permanganate embedding medium for electron microscopy . fixation of plant cells. J. Biophys . Biochem . J. Ultrastruct. Res . 26 :31 . Cytol . 6 :431 . 20. STEERE, R. L . 1957. Electron microscopy of 11 . MOLLENHAUER, H . H . 1964 . Plastic embedding structural detail in frozen biological specimens . mixtures for use in electron microscopy . Stain J. Biophys . Biochem . Cytol. 3 :45 . Technol. 39 :111 . 21 . STOECKENIUS, W ., and S . C. MAHR . 1965. Studies 12 . MOOR, H . 1965. Freeze-etching in high vacuum, on the reaction of osmium tetroxide with Report No. 2, Balzers A. G. Balzers, Princi- lipids and related compounds . Lab . Invest. 14 : pality of Liechtenstein . 458. 13 . MOOR, H., and K . MÜHLETHALER . 1963. Fine 22. TRUMP, B. F., and J . L . E. ERICSSON . 1965. The Downloaded from http://rupress.org/jcb/article-pdf/48/2/387/1069571/387.pdf by guest on 28 September 2021 structure of frozen-etched yeast cells . J. Cell effect of the fixative solution on the ultra- Biol. 17 :609 . structure of cells and tissues-a comparative 14 . MORRÉ, D. J., and H. H . MOLLENHAUER . 1969. analysis with particular attention to the proxi- Studies on the mechanism of glutaraldehyde mal convoluted tubule of the rat kidney . Lab . stabilization of cytomembranes. Proc. Indiana Invest. 14 :507. Acad . Sci. 78 :167 . 23 . VON HAHN, H . P. 1961. The specificity of osmium- 15 . PEASE, D . C. 1967 . Eutectic ethylene glycol and tetroxide as fixative and reagent in histochemis- pure propylene glycol as substituting media for try . In Bulletin of the Institute of Cellular the dehydration of frozen tissue . J. Ultra- Biology. The University of Connecticut, New struct. Res. 21 :75 . London, Conn. 3:1 . 16 . SABITINI, D . D ., K. BENSCH, and R. J . BARRNETT . 24. WEINSTEIN, R. S., and K . SOMEDA. 1967. The 1963. Cytochemistry and electron microscopy . freeze-cleave approach to the ultrastructure of The preservation of cellular ultrastructure and enzymatic activity by aldehyde fixation . J. frozen tissues . Cryobiology . 4 :116 . Cell Biol . 17 :19 . 25. WETZEL, B. K . 1961 . Sodium permanganate fixa- 17 . SABITINI, D . D ., J. MILLER, and R. J. BARRNETT . tion for electron microscopy . J. Biophys . Bio- 1964. Aldehyde fixation for morphological and chew. Cytol. 9 :711 .

3 9 4 THE JOURNAL OF CELL BIOLOGY . VOLUME 48, 1971