Elly van Donselaar, Department of Cell Biology, Institute of Biomembranes, University Medical Centre Utrecht, the NetherlandsSearch for more papers by this authorGeorge Posthuma, Department of Cell Biology, Institute of Biomembranes, University Medical Centre Utrecht, the NetherlandsSearch for more papers by this authorDagmar Zeuschner, Department of Cell Biology, Institute of Biomembranes, University Medical Centre Utrecht, the Netherlands Current address: Max-Planck-Institute of Molecular Biomedicine, Electron Microscopy, Roentgenstr. 20, 48149 Muenster, GermanySearch for more papers by this authorBruno M. Humbel, Cellular Architecture and Dynamics, Electron Microscopy and Structural Analysis, Utrecht University, Padualaan 8, 3584 CH Utrecht, the NetherlandsSearch for more papers by this authorJan W. Slot, Corresponding Author Department of Cell Biology, Institute of Biomembranes, University Medical Centre Utrecht, the NetherlandsDr Jan W. Slot, janwslot@hetnet.nlSearch for more papers by this author Elly van Donselaar, Department of Cell Biology, Institute of Biomembranes, University Medical Centre Utrecht, the NetherlandsSearch for more papers by this authorGeorge Posthuma, Department of Cell Biology, Institute of Biomembranes, University Medical Centre Utrecht, the NetherlandsSearch for more papers by this authorDagmar Zeuschner, Department of Cell Biology, Institute of Biomembranes, University Medical Centre Utrecht, the Netherlands Current address: Max-Planck-Institute of Molecular Biomedicine, Electron Microscopy, Roentgenstr. 20, 48149 Muenster, GermanySearch for more papers by this authorBruno M. Humbel, Cellular Architecture and Dynamics, Electron Microscopy and Structural Analysis, Utrecht University, Padualaan 8, 3584 CH Utrecht, the NetherlandsSearch for more papers by this authorJan W. Slot, Corresponding Author Department of Cell Biology, Institute of Biomembranes, University Medical Centre Utrecht, the NetherlandsDr Jan W. Slot, janwslot@hetnet.nlSearch for more papers by this author Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URLShare a linkShare onEmailFacebookTwitterLinked InRedditWechat Abstract Immunogold labeling of cryosections according to Tokuyasu (Tokuyasu KT. A technique for ultracyotomy of cell suspensions and tissues. J Cell Biol 1973;57:551–565), is an important and widely used method for immunoelectron microscopy. These sections are cut from material that is chemically fixed at room temperature (room temparature fixation, RTF). Lately in many morphological studies fast freezing followed by cryosubstitution fixation (CSF) is used instead of RTF. We have explored some new methods for applying immunogold labeling on cryosections from high-pressure frozen cells (HepG2 cells, primary chondrocytes) and tissues (cartilage and exocrine pancreas). As immunolabeling has to be carried out on thawed and stable sections, we explored two ways to achieve this: (1) The section fixation method, as briefly reported before (Liou W et al. Histochem Cell Biol 1996;106:41–58 and Möbius W et al. J Histochem Cytochem 2002;50:43–55.) in which cryosections from freshly frozen cells were stabilized in mixtures of sucrose and methyl cellulose and varying concentrations of glutaraldehyde, formaldehyde and uranyl acetate (UA). Only occasionally does this method reveal section areas with excellent cell preservation and negatively stained membranes like Tokuyasu sections of RTF material. (Liou et al.) (2) The rehydration method, a novel approach, in which CSF with glutaraldehyde and/or osmium tetroxide (OsO4) was followed by rehydration and cryosectioning as in the Tokuyasu method. Especially, the addition of UA and low concentrations of water to the CSF medium favored superb membrane contrast. Immunogold labeling was as efficient as with the Tokuyasu method. High-resolution immunoelectron microscopy has been, and still is, an important tool in cell biological research. Immunogold labeling on thin cryosections of chemically fixed cells is one of the most favorable protocols available. With this so-called Tokuyasu method, named after its originator 1, sensitive immunoreactions are achieved in non-resin-embedded sections of room temperature fixed (RTF) cells. Fixation is usually rather weak, with formaldehyde and/or low concentrations of glutaraldehyde. The preparations do not go through denaturizing processes like dehydration and embedding in resins. Furthermore, the small gold particles guarantee a good resolution, allow for quantification of the immunoreaction and make double labeling a simple option 2, 3. Finally, processing of sections after labeling 4 results in beautiful visualization of ultrastructure with an outstanding definition of cellular membranes. Over the years, we have used the Tokuyasu method in numerous localization studies and further improved it to a reliable routine technique 5-7. A concern is, however, that the time fixing chemicals take to completely penetrate into cells and tissues allows cellular processes to proceed for a short while that might result in structural and distributional artifacts. It has been shown that this time lag of fixation induces alterations of sensitive membrane structures 8-11. These artifacts would have important repercussions for immunocytochemical studies. For this reason we found it important to explore alternative fixation procedures. Nowadays many morphological studies, in particular, the ones set up to analyze cellular structures in three dimensions by electron tomography 12-15, start from cells that are structurally stabilized within a few milliseconds, by freezing. Fast freezing is facilitated by the development of high-pressure freezing (HPF) equipment and its refinement 16-21. The first choice in terms of optimal preservation would be imaging of the cells in the frozen–hydrated state 22, however, there is no labeling method yet for this approach. Therefore, fixation subsequent to freezing is needed to allow on-section immunogold labeling. Currently, this is mostly achieved by fixation with classical chemicals like glutaraldehyde or OsO4 during and after dehydration of the cryofixed material at −90°C in an organic solvent, e.g. acetone 23, 24, at 908C. This cryosubstitution fixation (CSF) is then physically supported by resin embedding 25. Sections of the embedded material can then be immunolabeled. However, compared to the non-embedded sections obtained by the Tokuyasu method, visualization of membranes in the resin sections is often poor and the efficiency of the immunolabeling is often much lower. One way to circumvent these disadvantages is by direct sectioning of the cryofixed material, thawing the sections in fixative and immunolabeling them. We described this section fixation method (SFM) briefly in an earlier report 5 and applied it successfully in a lipid localization study 26. Here, we describe a new method that combines the structural preservation by CSF with the high labeling efficiency of the Tokuyasu cryosectioning method. We conclude that for many questions in cell biology the less laborious and well-established Tokuyasu method keeps its full merits. However, in critical cases, we offer a reliable alternative, of which the potency has been exemplified in a recent publication of our group 27. Our aim was to combine the merits of cryofixation with those of the Tokuyasu procedure of cryosectioning and immunolabeling. In the last procedure, material is first chemically fixed, then immersed in a high concentration of sucrose before being frozen in liquid nitrogen and cryosectioned. Chemical fixation is a prerequisite for cryosections to maintain integrity during thawing and subsequent immunolabeling. Therefore, HPF material had first to be dehydrated, stabilized by fixation and rehydrated before entering the Tokuyasu procedure. The common way for chemical fixation after cryofixation is by CSF: Fixatives are added to the organic solvent in which the frozen material is dehydrated at low temperature 25. After CSF, the dehydrated samples are usually embedded in resin for sectioning and morphological observation or can be immunolabeled 28, 29. In general, resin embedding has a negative effect on labeling efficiency. To avoid that, we developed a rehydration procedure after CSF. To prevent re-crystallization in the vitreous frozen material, CSF was started below −90°C. After a relatively long period (48 h) at this temperature the samples were warmed by 2°C/h up to −30°C. To promote fixation, intervals of 8 h were inserted in the warming up at −60°C and at the end (−30°C) 25, 30. Hereafter the fixed material was rehydrated on ice and further processed in the same way as the Tokuyasu procedure (Figure 1). Flow chart of the different preparation procedures. Section fixation method (SFM), rehydration method (RHM) and standard Tokuyasu method. As a main model we used joint cartilage taken from the hip of young mice because this tissue has a mixed fibrillar and cellular composition with different densities, a challenge for obtaining good morphology. Furthermore, it can be handled relatively easily in a HPF machine and HPF has been applied successfully to cartilage before 19, 31, 32. To widen the scope of the study at certain points, we also used rat pancreatic tissue, HepG2 cells and human primary chondrocytes. In young mice, chondrocytes are relatively abundant in the joint cartilage. They occur as flattened fibroblast-like cells at the surface layer and differentiate into rounded cells in deeper layers. Initially the cytoplasm is largely occupied by the secretory structures rough endoplasmic reticulum (RER) and Golgi complex (Figure 2A), but when arriving at deeper layers where mineralization starts, these structures are mostly replaced by large glycogen deposits (Figure 2B). Our observations were mostly performed in the rounded superficial cells with an extensive secretory system and sometimes with developing glycogen deposits. Mouse cartilage, RTF and embedded in Epon. A) differentiated chondrocyte in surface layer. Apart from the nucleus, most of the cytoplasm is occupied by rough endoplasmic reticulum (ER) and Golgi structures (G). M, mitochondria. B) Chondrocyte in deeper layer of the cartilage, with large glycogen deposits (Gly). Note the atypical condensed nuclear chromatin, the extensive ER lumens and mitochondria (M) and the cellular debris (D) in the extracellular matrix. Scale bars represent 500 nm. We choose glutaraldehyde as a fixative during CSF because of its moderate effect on immunoreactivity. Cryosections of cartilage rehydrated after CSF with glutaraldehyde appear with a rather wrinkly collagen matrix (Figure 3A), where the cell profiles arise as nicely flattened. Higher magnifications showed dense cell contents in which most of the cellular membranes lacked contrast so that ultrastructural details were poorly visible (Figure 3B–D). For example the elements of the Golgi complex can only be identified because of well-fixed contents, but the membranous lining of the cisternae and vesicles is badly resolved (Figure 3B). Likewise, mitochondrial membranes did not show up clearly (Figure 3C). Mouse cartilage, HPF, CSF with glutaraldehyde and without (A, B, C), or with osmium tetroxide (D) and rehydration before cryosectioning. Immunogold labeling of SOD-I. A) At low magnification a chondrocyte with large glycogen deposits (Gly) in cytoplasm with well-developed ER (ER). Artificial distortions of the cryosections occur frequently in the extracellular matrix (arrows) but not in the cells. On higher magnification B) membranes of ER and Golgi complex (G) are poorly visible. C) ER and mitochondria (M) are defined by well-fixed contents and surroundings, but the membranes are rather obscure. At cross sectioned RER SOD-I labeling appears predominantly along the membranes (arrowheads). D) Osmium tetroxide in the CSF fixative slightly improved the membrane preservation. Small vesicles (arrowheads) and other membranes in the Golgi area (G) and of the nuclear envelope (arrow) are indicated. Scale bars represent 200 nm. The failure to visualize the membranes after HPF followed by CSF was not solved by extending the glutaraldehyde fixation after rehydration, neither the addition of OsO4 to the fixative nor replacement of glutaraldehyde by OsO4 during CSF gave a better image of the membranes (Figure 3D). In cryosections of RTF material, recognition of intracellular membranes is a strong feature, as illustrated for Golgi structures (Figure 4A), ER (Figure 4B), mitochondria (Figure 4C) and small vesicles in the extracellular matrix (Figure 4D). In the past, we have shown this for many tissues and cell types 5. Mouse cartilage, after Tokuyasu procedure, RTF with formaldehyde and glutaraldehyde, immunogold labeling of SOD-I. A) Membranes of Golgi structures (G), B) ER, C) mitochondria (M), D) and vesicles in the extracellular matrix are well preserved and visualized. SOD-I labeling is predominantly present along the ER membranes B). Note the loss of cytosolic material due to the weak chemical fixation A) arrows. Scale bars represent 200 nm. We have shown before that the SFM can result in excellent membrane contrast 5, 26. Simultaneous thawing and fixation is critical to the method. In practice only small parts of the sections are properly fixed, which makes this method unsuitable for routine immuno-electron microscopy. Nevertheless, in well-fixed cells the direct view of the HPF material, without dehydration, revealed very good preservation of membranes, as shown for the Golgi complex and ER in Figure 5. As in the standard Tokuyasu technique, membranes stood out in negative contrast. This shows that HPF did not, in itself, obscure cellular membranes in the sections of CSF treated tissue. Apparently, some change in the configuration of membranes occurred at a later stage of the procedure. Mouse cartilage, after SFM. Well-preserved membranes of ER as well as Golgi complex (G) with associated vesicles (asterisks) are visible in negative staining. Scale bar represents 200 nm. We next sought to solve the problem with two adjuvants in the CSF medium that might have a stabilizing and contrasting effect on the membranes: uranyl acetate (UA) and water. In studies on resin-embedded material, UA is added during CSF in many cryofixation studies 25. It has been shown that UA retains lipids, predominately phospholipids, during the substitution process 25, 33. UA is also used as a membrane-lipid stabilizing reagent in the standard Tokuyasu procedure 34, 35. In the SFM we found that UA is an obligatory component of the fixation drop (Figure 4) 5. Water, when added at low concentration during CSF, has been described to improve the membrane visibility significantly in resin-embedded tissues 36. HPF followed by CSF with UA in the CSF medium gave variable morphology in the cryosections. Especially in tissues like cartilage, the preservation of cellular membranes was unsatisfactory (Figure 6A and B). The bilayer of the RER became visible, but not the cristae of the mitochondria. The slight difference in contrast between the different cell constituents made it difficult to follow the membranes accurately. With exocrine pancreas we had similar results (not shown here). In cultured HepG2 cells membrane contrast was better. Membranes of the Golgi cisternae and adjacent ER elements were clearly depicted, as were the typical groups of vesicles and tubules present in the Golgi area (Figure 6C). In addition, cryosections of HepG2 allowed immunogold labeling of such critical structures as COPII vesicles (Figure 6C). Excellent membrane visibility was also obtained with human chondrocytes (Figure 6D). The quality of the contrast of these sections was in many cases as good as that of the routine Tokuyasu method. Thus, cryosections of cryofixed cultured cells that had been treated with UA-enriched medium during CSF showed a satisfactory membrane contrast. HPF followed by CSF in glutaraldehyde and UA, rehydrated before cryosectioning. A) Mouse cartilage, immunogold-labeled for COPII. B) Mouse cartilage, unlabeled. Membranes of the Golgi complex (G), ER and mitochondria (M) do not show up clearly in A) and B). C) HepG2 cell, labeled for COPII and D) cultured human chondrocyte, SOD-I labeled, same preparation procedure as in A) and B). The addition of uranyl ions improves membrane preservation in cultured cells considerably. G, Golgi complex; E, endosome; PM, plasma membrane; arrowheads pointing to COPII vesicles. Scale bars represent 200 nm. The addition of water during CSF also affected the membrane morphology after rehydration in tissue cells. Together with glutaraldehyde fixation there was some improvement (Figure 6B), but much better results were obtained when UA and water were added together during CSF. Both, in cartilage (Figure 7) and pancreas (Figure 8), the membranes were clearly visible and stood out as negatively stained lines. Together with prominent structures like RER, nuclear envelope, Golgi stacks and mitochondria, even the delicate transitional elements between ER and Golgi were often well visualized (Figures 7 and 8). Also membrane coats and filaments could be observed as nicely as in the best standard Tokuyasu cryosections. Mouse cartilage, HPF followed by CSF in glutaraldehyde, UA and 5% water, rehydrated before cryosectioning. Excellent membrane preservation of all compartments, immunolabeling of COPII. A) Stack of cisternae of the Golgi complex (G) with small COPII positive structures at its cis-side. N, nucleus. E, endosome. B) ER with budding profiles of COPII positive structures (arrowheads). M, Mitochondrion. Scale bars represent 200 nm. Rat exocrine pancreas, HPF followed by CSF as in Figure 7. A) COPII labeling marks the cis-Golgi elements in the periphery of a Golgi complex (G) between secretory granules (SG) and mitochondria (M). Immature secretory granules (ISG) are at the trans-side of the Golgi cisternae (G). B) Detail of an area like shown in A) Between well stained ER elements and Golgi cisternae (G) vesicles and amorphous material at the cis-side of the Golgi stack (G) are labeled for COPII. C) Unlabeled section showing the excellent preservation and visibility of the delicate structures (asterisk) at the cis-side of the Golgi stack (G). D) Low contrast area of a section immunolabeled for amylase. The labeling marks the lumen of the ER cisternae. M, mitochondria. Scale bars represent 200 nm. A special feature of the morphology was that in some sections areas, next to cellular membranes, cytoplasmic details were obscured by a general dense appearance of the cytoplasm causing a lack of differential contrast between these elements (Figure 9A). One factor controlling these variations may be the quality of freezing. Our impression is that the lack of contrast in the cytoplasm was more prominent when the freezing had been most successful. The quality of the freezing was judged from the appearance of the nuclear contents: ice crystal formation causes segregation of cellular components, the nucleus being the most sensitive structure. We observed no obvious differences between the dense and well-contrasted areas with respect to the immunoreactivity or ultrastructure. Chondrocytes, HPF followed by CSF, 2% UA and 5% water (A) or CSF with osmium tetroxide and 1% water (B, C), rehydrated before cryosectioning, immunolabeling of COPII. A) The cytosol is grayish and non-contrasted in which the membranes stand out in negative staining. B) and C) by adding osmium tetroxide during CSF, the resolution of fine structures improved. Scale bars represent 200 nm. OsO4 was used in an experiment with cartilage samples instead of glutaraldehyde for CSF in the presence of UA and water. The results were similar to the glutaraldehyde fixed sections. Sections from osmicated tissue had finer, better-delineated contrast (Figure 9B and C). We used three antibodies in our studies. One, anti-sec23, recognizes the COPII coats that are active in vesicle transport between ER and Golgi elements. The present results confirm our recent study on HEPG2 cells 27, which was supported by observations in the exocrine pancreas cell. The distribution of this membrane associated antigen in rehydration sections of the pancreas (Figure 9A and B) was similar to that described in Tokuyasu sections 37, 38. In rehydration sections of HepG2 cells (Figure 6C), cartilage (Figure 7) as well as exocrine pancreas (Figure 8A and B), we observed labeling associated with small vesicles, tubules and RER-associated buds. These labeled structures were located at the cis-side of the stack of Golgi cisternae and distant from the Golgi area between the ER lamellae. No obvious difference of COP II labeling was encountered after osmium tetroxide fixation (Figure 9B and C). The other two immunoreactions that we used were against soluble proteins. Amylase present in the vacuolar system and the cytosolic superoxide dimutase(SOD)-1. Amylase was strictly localized to the luminal space of the secretory compartments of pancreatic cells in rehydration sections (Figure 8D). SOD-1 labeling was performed in chondrocytes. Gold particles were found mainly in the cytosol between the ER cisternae and inside the nucleus. Between the ER cisternae the gold label was preferentially near the ER membranes, both in Tokuyasu and in rehydrated sections (Figure 4B and Figure 3B, respectively). We embarked on this study because of the growing idea that physical fixation is a better starting point for chemical fixation of cells and tissues. Our group and others have argued that the Tokuyasu procedure for cryosectioning of chemically fixed material is the best method for high efficiency immunolabeling 5-7. In this study we wished to combine the advantages of cryofixation with immunogold labeling of thawed ultra-thin cryosections. Cryosections suitable to immunolabel were obtained in two ways: they were cut directly from the frozen cells and chemically fixed during thawing (SFM), or the frozen material itself was first dehydrated by CSF, then rehydrated after which Tokuyasu cryosections were prepared (rehydration method, RHM). In most of our experiments, we used glutaraldehyde in the CSF medium. In some, we replaced glutaraldehyde by OsO4. Although the SFM is attractive because it seems straightforward and simple, for most studies SFM is unpractical because only small section areas have the desired morphological quality. This is probably due to the critical interweaving of thawing and chemical fixation. We therefore chose to develop the RHM. The main difficulty that we encountered here was that, after CSF with glutaraldehyde, membranes were hardly visible. An explanation can be that fixatives like glutaraldehyde and OsO4 only become reactive at temperatures higher than −90°C 25, 39, 40, which is the temperature at which the tissue is dehydrated. As a result, in the standard CSF procedure, the specimens are dehydrated before they are fixed. During that time they remain non-fixed in acetone. As the hydrophobic environment is then no longer restricted to their inner part, the molecular arrangement of membranes may become disturbed. When the temperature then goes up, the molecules will become fixed in disorder and the typical membrane structure is irreversibly lost. Another explanation is that because of the much better preservation of the cellular components, the lipids were not accessible for the heavy metal ions and therefore membranes could not be contrasted 33, 41, 42. We found two factors that counteracted loss of membrane visibility: addition of UA and low concentrations of water to the CSF medium. UA is used in many procedures as an additional fixative for electron microscopy. In Tokuyasu sections it is used to provide the final contrast, but also, at neutral pH, to stabilize membranes and prevent them from being extracted from the sections 34. UA treatment is scheduled late in the Tokuyasu procedure, possibly to minimize the exposure of the sections to an agent that could harm the immunoreactivity. Previously, we have observed a much stronger effect of UA on these sections by adding it to the pick-up solution in which Tokuyasu sections are thawed. In particular when they were immediately dried in UA/methyl cellulose the typical negative membrane contrast was enhanced 5, but also when UA was washed away before and during relatively long immuno-incubations, the improvements remained for a great part 6. UA is an obligatory ingredient during the SF procedure 5 (Figure 5). Hence, in aqueous milieu UA stabilizes membranes. Uranium contrast in cryosections is determined by unreactive, unstained membranes amidst positively uranium-stained cytoplasmic material, mainly proteins. In favorable Tokuyasu sections cytoplasmic elements like ribosomes, microtubules and filaments stand out distinctly and membrane coats can be seen on Golgi elements and the plasma membrane. Under less favorable conditions, however, the cytoplasm is dense and homogeneously stained and except for the membranes, cytoplasmic details are not clearly visible. This happens in particular after stronger fixations in glutaraldehyde. Formaldehyde fixation, on the other hand, results in much more differentiated cytoplasmic contrast. The dense homogeneous cytoplasm was also revealed in native cryosections, which were picked up in a UA containing solution 6. Our explanation is that in the favorable sections a sort of differential extraction of cytoplasmic components occurs after thawing and during the long immunolabeling procedure. Strong fixation and, apparently, UA counteracts such extraction, which results in a brilliant visualization of membranes in an otherwise poorly stained cytoplasmic background. In cryosections of rehydrated tissue, after CSF in the presence of UA, membrane contrast was largely lost. This was much less so in similar preparations of cultured cells. We hypothesize that apart from the fixation conditions, the natural density of the cytoplasmic proteins plays a role in membrane contrast formation also. The addition of low concentrations of water during CSF for enhancing membrane contrast has been introduced by Walther and Ziegler in Epon studies 36. Here we found a similar effect during the rehydration procedure. In particular membranes of the ER and Golgi complex were well preserved and detectable by the typical negative uranium contrast in the thawed cryosections. It was argued before that formation of small ice crystals that nucleate along the membranes could have induced the membrane lining 36. That is not unlikely as acetone at −90°C is saturated with water below 1% 28, 43. That means that cryosubstitution will be delayed until the temperature goes up. The vitreous ice in which the cells are initially frozen may re-crystallize in relatively small invisible ice crystals that make small ‘holes’. These ‘holes’ allow heavy metal ions to access and bind the negatively charged lipids, hence contrasting the membranes. Another possible explanation for the improvement of membrane preservation is that the delay in substitution brings about a much tighter connection between dehydration and fixation so that membranes do not become disorganized before fixation as in the standard procedure. Finally, it may be possible that the small amount of water added changes the reaction properties of the chemical fixatives. This may also help maintaining and contrasting the membrane structures. The aim of our study is to present the RHM as an alternative procedure that can be used to combine the superior preservation by cryoimmobilization, e.g. by HPF, with the excellent immunolabeling properties of Tokuyasu cryosections. CSF has the advantage of starting from cryoimmobilized tissues or cells, which is believed to be closest to the living state. Then the cells are immobilized in their actual metabolic state in the natural surrounding and no morphological changes due to loss of ions, changed osmolarity, ‘switching off’ of membrane pumps and so on is to be expected. This CSF method has, however, also disadvantages: Firstly, the cells or tissue have to pass through a dehydration step with possible denaturizing effects, as we observed for the membranes. Secondly, HPF samples are small and often the freezing quality varies. Thirdly, the tissue is vulnerable to manipulation before the actual cryoimmobilization. Studies from plastic-embedded CSF versus RTF material have shown that structural differences occur 8-10, 44-47. On the other hand, in our sections of different tissues, the general appearance of the cellular ultrastructure after RHM looked similar to what we are used to see in Tokuyasu sections. This was confirmed in a special study on the transitional structures between ER and Golgi complex 27. This center of presumably high membrane dynamics did not seem to be affected differently by either method. Also we observed in both procedures three antibody reactions against a membrane associated a cytosolic protein and a secretory protein. The labeling pattern and efficiency was the same for CSF rehydrated and Tokuyasu sections. We noted that osmium tetroxide added to the substitution medium did not negatively influence the efficiency of immunolabeling. The label density was comparable to preparations without osmium tetroxide. This result might reflect the fact that the chemical reaction of osmium tetroxide at subzero temperature is different, mainly fixing 39, than at temperatures above zero, were the action is mainly proteolytic 48-50. In summary, here we describe an excellent method to combine the preservation of the natural cellular ultrastructure by HPF with the high efficient immunogold labeling of Tokuyasu cryosectioning. Our results suggest that this method will be of great value especially for difficult to fix biological material 51, e.g. plants 47, 52, fungi 53, nematodes 54, 55 and delicate membrane interactions 10, 11. The results, however, also confirm the high quality of the standard Tokuyasu procedure. We suggest using the standard Tokuyasu technique whenever possible because it is much easier and faster and allows the preparation of larger samples. In the present study, we have used three methods to prepare non-resin-embedded, ultra-thin cryosections for immunogold labeling (Figure 1). SFM has only been used in a few studies so far 5, 26. It starts with HPF material that is directly cryosectioned at −160°C and the sections are subsequently picked up with a fixative-containing solution for fixation during thawing. RHM is new and, like the first method, starts with HPF material, but is followed by dehydration via cryosubstitution in acetone. Fixation occurs at low temperature under non-aqueous conditions 25. After substitution the material is rehydrated and further processed for cryosectioning, like described in RTF method. RTF method according to Tokuyasu is refined over several decades of practice and developed into a routine method. This approach starts with chemical fixation of biological material at room temperature. High concentrations of sucrose are infused into fixed material, allowing simple freezing in liquid nitrogen without ice damage. The material is subsequently cryosectioned 1, 5, 56, 57. Hyaline cartilage samples were collected from the hip joint of 5-week-old mice. The animals were killed by neck dislocation and the femur was removed. 0.2-mm thick slices were cut tangentially from the convex joint surface and immersed in cell culture medium (DMEM, Gibco, Invitrogen Corporation, Inchinnan Business Park, Paisley, UK). Slice fragments of about 1 mm2 were immersed in 1% gelatin in cell culture medium and put into the flat specimen carrier (0.2 mm deep and 1.2 mm in diameter) of the HPF apparatus 20 (EM-PACT-1, Leica Microsystems, Vienna, Austria). Cryoimmobilization at a pressure of 2000 bar (2 × 108 Pa) was performed according to the manufacturer\'s manual within 1 min after sampling of the tissue. Pancreas biopsies were taken from anaesthetized Wistar rats (200 g, approximately 6 weeks old), using the HPF microbiopsy system 58 (Microbiospy Transfer System; Leica Microsystems). The biopsies were immediately dipped in 1-hexadecene 59 and transferred from the needle to the flat specimen carrier with the special transfer device according to the manufacturer\'s manual. Cryoimmobilization occurred within 30 s after the biopsy was taken. HepG2 cells were cultured on Cytodex 3 beads (Amersham Biosciences, Roosendaal, the Netherlands) to almost 100% confluence 27. Then the beads were mixed with 1% gelatin in DMEM medium, transferred to the flat specimen carrier and cryoimmobilized as above. Primary cultures of human chondrocytes were grown on collagen II-coated filters (Millicell-CM, 0.4 μm culture plate insert, 12 mm; Millipore, Carrigtwohill, Cork, Ireland). Pieces of 1 mm2 of these cultures were excised and fitted into the flat specimen carrier of the HPF apparatus. Cryosectioning of freshly frozen cartilage was performed as described by Studer et al. 19. The frozen cartilage samples were removed from the specimen holders in the cryochamber of an cryo-ultramicrotome, set at −160°C. Then the fragments were glued to a microtome specimen holder with a drop of a viscous mixture of 2-propanol/ethanol (3/2), at −145°C that hardens while temperature was again lowered to −160°C. Trimming and sectioning were carried out at −160°C with a dry diamond knife (Element Six B.V., Cuijk, the Netherlands or Diatome AG, Biel, Switzerland). Flat and ultra-thin (∼80 nm) sections were shifted away from the knife edge with an eyelash. A loop with 3 mm diameter made from a wire of stainless steel (0.3 mm thickness) was filled with a droplet of 60 mM PIPES, 25 mM HEPES, 10 mM EGTA and 2 mM MgCl2, pH 6.9 (PHEM buffer). 60 containing 1% (w/v) methyl cellulose, 0.25% (w/v) UA, 0.2% (v/v) glutaraldehyde and 2% (w/v) formaldehyde in PHEM buffer. After entering the cryochamber, the loop, mounted on a wooden stick, was brought to the sections as quick as possible. The droplet was pushed on the sections so that these adhered as flat as possible to the surface. The initiation of fixation and spreading of the sections over the droplet\'s surface occurred after the loop was withdrawn from the cryochamber and the droplets started to thaw. After thawing, the sections floating on the down side of the droplet were placed on Formvar carbon-coated copper grids so that they attached to the Formvar film while covered with the pick-up droplet. This situation was kept at room temperature for 10 min to finish chemical fixation. Then the grids were washed by floating for 10 min on drops of 1.8% methyl cellulose and 0.4% UA and dried in a thin film of that solution for direct EM observation. Alternatively, the sections were transferred to drops of buffered washing solutions for immunogold labeling (see below). Phosphate-containing buffers were avoided as phosphate tends to precipitate uranyl ions, even after thorough washing. The cryofixed samples were transferred to the acetone-based substitution medium in 1.5 mL microtubes placed at −90°C in a cryosubstitution apparatus (AFS, Leica Microsystems). To the acetone different additives (UA, osmium tetroxide, glutaraldehyde, water) in different combinations were added (details are described below and illustrated in Table 1). The samples were dehydrated and fixed in this solution according to the following protocol: −30°C/4 rinses in cryosubstitution-solution w.o. UA (!) on ice/1 h Four rinses in acetone Kept at −30°C for 8 h. When the solution contained UA it was washed off by four times rinsing with the same substitution medium but without UA. The tubes were removed from the substitution apparatus and placed for 1 h on ice. For chemical fixation we added either 0.5% glutaraldehyde or 1–2% OsO4 (step 6) to the acetone until the washing. Sometimes 0.1–0.2% UA was added during CSF until step 5 and sometimes 1–5% water was added to the substitution medium throughout the entire procedure. After the dehydration and fixation during cryosubstitution, the samples were rehydrated on ice in six steps of 10 min each in 95, 90, 80 and 70% (v/v) acetone in distilled water, then 50% (v/v) acetone in PHEM buffer and finally in 30% acetone in PHEM buffer. Then the samples were washed four times for 10 min in PHEM buffer. In case of glutaraldehyde fixation GA was present during rehydration and washed away in PHEM buffer. Finally, the samples were immersed for 10 min at 37°C in 12% gelatin in PHEM buffer. After gelation at 4°C small blocks were trimmed for sucrose infusion and cryosectioning as described for the Tokuyasu procedure (see below). The only difference is that, as in the section fixation procedure, phosphate-containing buffers were avoided at any stage. Hyaline cartilage samples were collected as described above and fragments of about 1 mm2 were selected from the slices and chemically fixed by immersion in 2% (w/v) formaldehyde, 0.2% (v/v) glutaraldehyde in 0.1 m sodium phosphate buffer, pH 7.4 (PB) for 2 h at room temperature. Then the fragments were washed a few times in PBS and transferred to 12% gelatin in PBS at 37°C for 10 min. HepG2 cells were grown to 100% confluence in culture dishes (60 mm × 15 mm; Corning Inc, Corning, MA, USA). Double-strength fixative (4% (w/v) formaldehyde, 0.4% (v/v) glutaraldehyde) in PB was added to the plates in an equal volume to the culture medium, while swirling gently, so that it was diluted immediately to standard strength (2% (w/v) formaldehyde, 0.2% (w/v) glutaraldehyde). After mixing, the fixative was replaced by fresh standard strength fixative and fixation proceeded for at least 120 min at room temperature. Then the cells were washed several times with PBS, scraped from the plate in PBS containing 1% (w/v) gelatin and pelleted at 37°C in 12% gelatin in PBS. Alternatively, HepG2 cells grown on Cytodex 3 beads, were immersed in fixative and likewise embedded in 12% gelatin blocks. Rat pancreas was perfusion fixed with 2% (w/v) formaldehyde, 0.2% (v/v) glutaraldehyde in PB as described 37, 38. Small fragments encapsulated in 12% gelatin for 10 min at 37°C. After gelation at 4°C convenient blocks (about 1 mm3) for cryo-ultramicrotomy were trimmed under a dissection microscope at 4°C. The gelatin-embedded blocks were immersed overnight in 2.3 m sucrose in rotating vials at 4°C. Then they were mounted on specimen holders of an ultramicrotome (UC6 or UCT, Leica Microsystems) and frozen by plunging into liquid nitrogen or placing into the cryochamber of the microtome. After trimming to suitable block shape, ultra-thin sections (50–60 nm) were cut at −120°C on dry diamond knives (Element Six B.V. or Diatome AG). Flat ribbons of glossy-looking sections were shifted from the knife edge with an eyelash, when needed under control of an ionizer (Diatome AG). They were picked up in a wire loop filled with a drop of 1% (w/v) methyl cellulose, 1.15 m sucrose in PHEM buffer. The sections were thawed on the pick-up droplet and transferred, sections downwards, to Formvar carbon-coated copper grids as described in the section fixation paragraph. Antibodies used were affinity purified rabbit polyclonals directed against: Cu–Zn superoxidase (type I), SOD-I, purified from rat liver (provided by Dr J. D. Crapo, Denver, CO); amylase isolated from rat pancreas 61 COPII coat protein sec23 (Affinity Bioreagents, Inc, Breda, the Netherlands). Protein A/gold conjugates was prepared according to the method described by Slot and Geuze 3. Sections on copper grids from either the rehydration or the Tokuyasu procedure can be stored for long time in the cold 62 still covered by the pick-up droplet. When immunolabeling starts the grids are placed on buffer at 37°C for 30 min to let the pick-up solution diffuse away together with the 12% gelatin. Next the grids are passed over a series of droplets of washing, blocking, antibody and protein A-gold solutions for routine labeling procedures 63. After a final wash in distilled water the sections are left for 5 min on 2% uranyl oxalate (pH 7) 34 and transferred, via a few seconds on a puddle of distilled water, to a mixture of 1.8% methyl cellulose and 0.4% UA. After 5–10 min the grids are looped out, most of the excess of the viscous solution was drained away and the sections are dried in the thin film that remains over the grid 4. The sections were viewed in a JEOL 1010 (JEOL, Japan) or 1200 electron microscope and images were recorded on Kodak 4489 (Kodak, NY, USA) sheet films. We thank Marc van Peski and Rene Scriwanek for assistance with the preparation of the figures. We thank Hans Geuze for stimulating discussions. Primary cultures of human chondrocytes were kindly provided by Drs L. Creemers and G. Auw Yang. Antibodies used were affinity purified rabbit polyclonals directed against: Cu–Zn superoxidase (type I) purified from rat liver (provided by Dr J. D. Crapo, Denver, CO). A part of this work was supported by the Cyttron Consortium and the European 3DEM Network of Excellence.Scheffer RC. Use of colloidal gold particles in double-labeling immunoelectron microscopy of ultrathin frozen tissue sections. J Cell Biol 1981; 89: 653– 665. A new method of preparing gold probes for multiple-labeling cytochemistry. Eur J Cell Biol 1985; 38: 87– 93. Griffiths G. Selective contrast for electron microscopy using thawed frozen sections and immunocytochemistry. In: JP Revel, T Barnard, GH Haggis, editors. Science of Biological Specimen Preparation, 1983. AMF O\'Hare, IL 60666-0507, USA: SEM Inc.; 1984. pp. 153– 159. Improving structural integrity of cryosections for immunogold labeling. Histochem Cell Biol 1996; 106: 41– 58.Slot JW. The autophagic and endocytic pathways converge at the nascent autophagic vacuoles. J Cell Biol 1997; 136: 61– 70. Immunogoldlabeling of ultrathin cryosections: application in immunology. Handbook of Experimental Immunology 1997; 4: 1– 11.Müller M. Light damage in rod outer segments: the effects of fixation on ultrastructural alterations. Current Eye Research 1996; 15: 807– 814.Pavelka M. Cryopreparation provides new insight into the effects of brefeldin A on the structure of the HepG2 Golgi apparatus. J Struct Biol 2000; 130: 63– 72.Baumeister W. Macromolecular architecture in eukaryotic cells visualized by cryoelectron tomography. Science 2002; 298: 1209– 1213.Howell KE. Direct continuities between cisternae at different levels of the Golgi complex in glucose-stimulated mouse islet beta cells. Proc Natl Acad Sci USA 2004; 101: 5565– 5570.Kleijmeer MJ. 3-D Structure of Multilaminar Lysosomes in Antigen Presenting Cells Reveals Trapping of MHC II on the Internal Membranes. Traffic 2004; 5: 936– 945.Humbel BM. Electron microscopy tomography and localization of proteins and macromolecular complexes in cells. In: EA Golemis, PD Adams, editors. Protein-Protein Interactions A Molecular Cloning Manual. New York: Cold Spring Harbor Laboratories Press; 2006. pp. 715– 739. Riehle U. Über die Vitrifizierung von verdünnter wässriger Lösungen. Zürich: Federal Institute of Technology (ETH); 1968. Caveolin-1 is not essential for biosynthetic apical membrane transport. Mol Cell Biol 2005; 25: 10087– 10096.Crowell J. Electron microscopy after rapid freezing on a metal surface and substitution fixation. Anat Rec 1964; 149: 381– 386. Steinbrecht RA. Cryofixation without cryoprotectants. Freeze substitution and freeze etching of an insect olfactory receptor. Tissue and Cell 1980; 12: 73– 100. Freeze-substitution for immunochemistry. In: AJ Verkleij, JLM Leunissen., editors. Immuno-Gold Labeling in Cell Biology. Boca Raton: CRC Press; 1989. pp. 115– 134.Slot JW. Immunoelectron microscopic localization of cholesterol using biotinylated and non-cytolytic perfringolysin O. J Histochem Cytochem 2002; 50: 43– 55.Klumperman J. Immuno-electron tomography of ER exit sites reveals the existence of free COPII-coated transport carriers. Nat Cell Biol 2006; 8: 377– 383.Müller M. Improved structural preservation by combining freeze substitution and low temperature embedding. Beitr Elektronenmikroskop Direktabb Oberfl 1983; 16: 585– 594. P Röhlich, D Szabo, editors. Proc 8th Eur Congr Electron Microsc. Budapest; 1984. pp. 1789– 1798. Improved structural preservation by freeze substitution. In: P Brederoo, W de Priester, editors. Proc 7th Eur Congr Electron Microsc. The Hague; 1980. pp. 720– 721.McDonald KL. The ultrastructure of the connective tissue matrix of skin and cartilage after high-pressure freezing and freeze-substitution. J Histochem Cytochem 1993; 41: 1141– 1153. Ultrastructure of hyaline cartilage. Acta Pathol Microbiol Immunol Scand Sect A Pathol 1986; 94: 313– 323.Müller M. ‘Lipidic particle’ systems as visualized by thin-section electron microscopy. Biochim Biophys Acta 1985; 812: 591– 495.Ziegler A. Freeze substitution of high-pressure frozen samples: the visibility of biological membranes is improved when the substitution medium contains water. J Microsc 2002; 208: 3– 10.Klumperman J. Vesicular tubular clusters between the ER and Golgi mediate concentration of soluble secretory proteins by exclusion from COPI-coated vesicles. Cell 1999; 98: 81– 90.Barrnett RJ. The chemical nature of osmium tetroxide fixation and staining of membranes by x-ray photoelectron spectroscopy. Biochim Biophys Acta 1976; 436: 577– 592. A Boyde, JJ Wolosewick, editors. The Science of Biological Specimen Preparation 1985. AMF O\'Hare: SEM Inc.; 1986. pp. 175– 183.Christiansson A. Extraction of proteins and membrane lipids during low temperature embedding of biological material for electron microscopy. J Microsc 1986; 142: 79– 86.Carlemalm E. Extraction of lipids during freeze-substitution of Acholeplasma laidlawii-cells for electron microscopy. J Microsc 1984; 134: 213– 216.Karnovsky MJ. Rapid-freezing cytochemistry: preservation of tubular lysosomes and enzyme activity. J Histochem Cytochem 1991; 39: 787– 792.Humbel BM. Influence of aldehyde fixation on the morphology of endosomes and lysosomes: quatitative analysis and electron tomography. J Microsc 2003; 212: 81– 90.Walther P. Comparison of ultrastructure of germinating pea leaves prepared by high-pressure freezing-freeze substitution and conventional chemical fixation. J Electron Microsc 1995; 44: 104– 109.Müller M. High-pressure freezing of soybean nodules leads to an improved preservation of ultrastructure. Planta 1992; 188: 155– 163. T Barnard, GH Haggis, editors. The Science of Biological Specimen Preparation 1983. AMF O\'Hare, IL 60666: SEM Inc.; 1983. pp. 1– 5.Pollard TD. Improved preservation and staining of HeLa cell actin filaments, clatrin-coated membranes, and other cytoplasmic structures by tannic acid-glutaraldehyde-saponin fixation. J Cell Biol 1983; 96: 51– 62.Mitsushima A. A preparation method for observing intracellular structures by scanning electron microscopy. J Microsc 1984; 133: 213– 222.Morphew MK. Improved preservation of ultrastructure in difficult-to-fix organisms by high pressure freezing and freeze substitution: I. Drosophila melanogaster and Strongylocentrotus purpuratus embryos. Microsc Res Tech 1993; 24: 465– 473. Hess MW. Of plants and other pets: practical aspects of freeze-substitution and resin embedding. J Microsc 2003; 212: 44– 52.Boekhout T. Automated electron tomography of the septal pore cap in Rhizoctonia solani. J Struct Biol 2000; 131: 10– 18.McDonald KL. Cryoimmobilization and three-dimensional visualization of C. elegans ultrastructure. J Microsc 2003; 212: 71– 80.Geuze HJ. Gold markers for single and double immunolabelling of ultrathin cryosections. In: JM Polak, IM Varndell, editors. Immunolabelling for Electron Microscopy. Amsterdam: Elsevier; 1984. pp. 129– 142.Porter KR. Stabilization of the cytoplasmic ground substance in detergent-opened cells and a structural and biochemical analysis of its composition. Proc Natl Acad Sci USA 1981; 78: 4329– 4333.Tokuyasu KT. Immunocytochemical localization of amylase and chymotrypsinogen in the exocrine pancreatic cell with special attention to the Golgi complex. J Cell Biol 1979; 82: 697– 707.Posthuma G. A reliable and convenient method to store ultrathin thawed cryosections prior to immunolabeling. J Histochem Cytochem 2002; 50: 57– 62.James DE. Immuno-localization of the insulin regulatable glucose transporter in Brown adipose tissue of the rat. J Cell Biol 1991; 113: 123– 135. The full text of this article hosted at iucr.org is unavailable due to technical difficulties. Please check your email for instructions on resetting your password. If you do not receive an email within 10 minutes, your email address may not be registered, and you may need to create a new Wiley Online Library account. Can\'t sign in? Forgot your username? Enter your email address below and we will send you your username If the address matches an existing account you will receive an email with instructions to retrieve your username