An improved procedure for low-temperature embedding of high-pressure frozen and freeze-substituted plant tissues resulting in excellent structural preservation and contrast

Here we describe refinements in the processing of high-pressure frozen samples of delicate plant tissues for immuno-electron microscopy. These involve: shortened freeze-substitution schedules, lower temperatures during processing and polymerisation, the avoidance of temperature fluctuations and the optimisation of heat transfer from the specimens using small disposable aluminium containers. The application of these modifications leads to very good structural preservation and selective membrane contrast. As a result, the versatility of the method is increased since not only immuno-electron microscopical studies can be performed but often the quality is also quite suitable for structural investigations.     

Controlled microaspiration for high-pressure freezing: a new method for ultrastructural preservation of fragile and sparse tissues for TEM and electron tomography

High‐pressure freezing is the preferred method to prepare thick biological specimens for ultrastructural studies. However, the advantages obtained by this method often prove unattainable for samples that are difficult to handle during the freezing and substitution protocols. Delicate and sparse samples are difficult to manipulate and maintain intact throughout the sequence of freezing, infiltration, embedding and final orientation for sectioning and subsequent transmission electron microscopy. An established approach to surmount these difficulties is the use of cellulose microdialysis tubing to transport the sample. With an inner diameter of 200 μm, the tubing protects small and fragile samples within the thickness constraints of high‐pressure freezing, and the tube ends can be sealed to avoid loss of sample. Importantly, the transparency of the tubing allows optical study of the specimen at different steps in the process. Here, we describe the use of a micromanipulator and microinjection apparatus to handle and position delicate specimens within the tubing. We report two biologically significant examples that benefit from this approach, 3D cultures of mammary epithelial cells and cochlear outer hair cells. We illustrate the potential for correlative light and electron microscopy as well as electron tomography.     

Candida albicans biofilms: comparative analysis of room‐temperature and cryofixation for scanning electron microscopy

Biofilms are frequently related to invasive fungal infections and are reported to be more resistant to antifungal drugs than planktonic cells. The structural complexity of the biofilm as well as the presence of a polymeric extracellular matrix (ECM) is thought to be associated with this resistant behavior. Scanning electron microscopy (SEM) after room temperature glutaraldehyde‐based fixation, have been used to study fungal biofilm structure and drug susceptibility but they usually fail to preserve the ECM and, therefore, are not an optimised methodology to understand the complexity of the fungal biofilm. Thus, in this work, we propose a comparative analysis of room‐temperature and cryofixation/freeze substitution of Candida albicans biofilms for SEM observation. Our experiments showed that room‐temperature fixative protocols using glutaraldehyde and osmium tetroxide prior to alcohol dehydration led to a complete extraction of the polymeric ECM of biofilms. ECM from fixative and alcohol solutions were recovered after all processing steps and these structures were characterised by biochemistry assays, transmission electron microscopy and mass spectrometry. Cryofixation techniques followed by freeze‐substitution lead to a great preservation of both ECM structure and C. albicans biofilm cells, allowing the visualisation of a more reliable biofilm structure. These findings reinforce that cryofixation should be the indicated method for SEM sample preparation to study fungal biofilms as it allows the visualisation of the EMC and the exploration of the biofilm structure to its fullest, as its structural/functional role in interaction with host cells, other pathogens and for drug resistance assays.     

Freeze-substitution: the addition of water to polar solvents enhances the retention of structure and acts at temperatures around –60°C

High‐pressure freezing followed by freeze substitution and plastic embedding is becoming a more widely used method for TEM sample preparation. Here, we have investigated the influence of solvents, fixative concentrations and water content in the substitution medium on the sample quality of high‐pressure frozen, freeze‐substituted and plastic embedded mammalian cell culture monolayers. We found that the visibility of structural details was optimal with acetone and that extraction increased with both increasing and decreasing solvent polarity. Interestingly, the addition of water to polar solvents increased the sample quality, while being destructive when added to apolar solvents. The positive effect of water addition is saturable in acetone and ethanol at 5%(v/v), but even addition of up to 20% water has no negative effect on the sample structure. Therefore, a medium based on acetone containing fixatives and 5% water is most optimal for the substitution of mammalian cell cultures. In addition, our results suggest that the presence of water is critical for the retention of structure at temperatures around –60°C.     

A review of high-pressure freezing preparation techniques for correlative light and electron microscopy of the same cells and tissues

In this paper, we review some published studies using correlative light and electron microscopy methods. We further refined our criteria to include only those studies using live cells for light microscope and where high-pressure freezing was the method of specimen preparation for electron microscopy. High-pressure freezing is especially important for some difficult-to-fix samples, and for optimal preservation of ultrastructure in samples larger than a few micrometres. How the light microscope observations are done is completely sample dependent, but the choice of high-pressure freezer depends on the speed required to capture (freeze) the biological event of interest. For events requiring high time resolution (in the 4–5 s range) the Leica EM PACT2 with rapid transfer system works well. For correlative work on structures of interest that are either non-motile or moving slowly (minutes rather than seconds), any make of high-pressure freezer will work. We also report on some efforts to improve the capabilities of the Leica EM PACT2 rapid transfer system.     

A new approach for high-pressure freezing of primary culture cells: the fine structure and stimulation-associated transformation of cultured rabbit gastric parietal cells

A newly designated procedure for high‐pressure freezing of primary culture cells provided excellent ultrastructure of rabbit gastric parietal cells. The isolated parietal cells were cultivated on Matrigel‐coated aluminium plates for conventional subsequential cryoimmobilization by high‐pressure freezing. The ultrastructure of different organelles (Golgi apparatus, mitochondria, multivesicular bodies, etc.) was well preserved compared to conventional chemical fixation. In detail, actin filaments were clearly shown within the microvilli and the subapical cytoplasm. Another striking finding on the cytoskeleton system is the abundance of microtubules among the tubulovesicles. Interestingly, some microtubules appeared to be associating with tubulovesicles. A large number of electron‐dense coated pits and vesicles were observed around the apical membrane vacuoles in cimetidine‐treated resting parietal cells, consistent with an active membrane uptake in the resting state. Immunogold labelling of H⁺/K⁺‐ATPase was seen on the tubulovesicular membranes. When stimulated with histamine, the cultured parietal cells undergo morphological transformation, resulting in great expansion of apical membrane vacuoles. Immunogold labelling of H⁺/K⁺‐ATPase was present not only on the microvilli of expanded apical plasma membrane vacuoles but also in the electron‐dense coated pits. The present findings provide a clue to vesicular membrane trafficking in cultured gastric parietal cells, and assure the utility of the new procedure for high‐pressure freezing of primary culture cells.     

Freeze substitution of high-pressure frozen samples: the visibility of biological membranes is improved when the substitution medium contains water

Biological membranes are often poorly visible with the electron microscope after high‐pressure freezing and freeze‐substitution. The water content of the sample and of the substitution medium is one factor among others that strongly influences membrane visibility. In order to investigate this effect, high‐pressure frozen yeast cells, rat‐pancreas tissue and arthropod tissue were freeze‐substituted with and without adding water to the substitution medium. The visibility of the biological membranes was generally improved if the substitution medium contained 1–5% water. The effect was especially pronounced in yeast cells, where membrane visibility was poor after freeze‐substitution with water‐free medium but good after addition of 5% water to the substitution medium.     

Ultrastructure of the frog retina after high-pressure freezing and freeze substitution

In many types of tissue, high-pressure freezing (HPF), followed by freeze substitution, can produce excellent ultrastructural preservation at depths over 10 times that obtained by other cryofixation techniques. However, in the case of neural tissue, the benefits of HPF have not been realized. In the present study, isolated frog ( Rana pipiens) retina was sliced at a thickness of 150 or 350 μm, rapidly frozen in a Balzers HPM 010 high-pressure freezer, and freeze substituted with 1% OsO₄ and 0.1% tannic acid in acetone. Specially designed HPF chambers and specific freezing media (35% high-MW dextran for 150-μm slices or 15% low-MW dextran for 350-μm slices) were required for adequate freezing.
The quality of preservation after HPF was excellent throughout the retina in both the 150- and 350-μm slices, compared with chemically fixed slices. Specifically, HPF resulted in better preserved cellular, mitochondrial and nuclear membranes in all retinal layers.
This is the first study to successfully cryofix all of the layers of the retina. The increased depths of adequate freezing achieved by HPF should facilitate various ultrastructural studies of retina, as well as of other CNS tissues, where preservation approaching that of the 'native' state is required.     

Chloroplast ultrastructure in leaves of Urtica dioica L. analyzed after high-pressure freezing and freeze-substitution and compared with conventional fixation followed by room temperature dehydration

In this article, we report on the adaptation of high-pressure freezing and freeze-substitution (HPF-FS) for ultrastructural analysis of leaf tissue with special emphasis on chloroplasts. To replace the gas in the intercellular spaces, a mixture of water and methanol (MeOH) was employed. We compared three different supplements for FS--osmiumtetroxide, uranyl acetate, and safranin--with regard to the preservation of the ultrastructure of chloroplasts and other cellular compartments. The results show that (i) replacement of air within intercellular spaces by 8% (v/v) MeOH has no influence on the ultrastructure of the chloroplasts, (ii) undulation of membranes frequently observed after conventional preparation of specimens does not occur during chemical fixation but during room temperature dehydration, and (iii) uranyl acetate or osmium tetroxide employed during FS are not superior over safranin.     

Towards more precise definition of conditions for satisfactory deep etching

Experiments were carried out to determine whether propane-jet freezing could be as satisfactory as impact freezing in deep etch work. The material used was the intestinal brush border, previously studied by Heuser and coworkers where the 5–6 nm decoration of actin rootlet filaments, and the fine network of filaments linking these rootlets, provide good criteria by which to judge the quality of the preparation, as regards ice crystal growth and surface contamination. Propane jet freezing was indeed found satisfactory provided appropriate conditions were met (viz thin specimen, fracture near the surface). Variable results were obtained until it was realized that with a Balzers freeze-etch unit fitted with a rotating specimen table there is a 10–15 min delay (when specimen temperature is reset) between the time the recording thermocouple shows a given temperature to have been obtained and the time the specimen block actually reaches this temperature. Appropriate allowance must be made for this lag to achieve satisfactory deep etch replicas.     

Comparison of polyethylene glycol and diethylene glycol distearate embedding methods for the preservation of fungal cytoskeletons

Hyphae of the fungus Basidiobolus magnus were embedded in extractable polyethylene glycol (PEG) or diethylene glycol distearate (DGD) to prepare embedment-free sections in order to seek components of the cytoskeleton that may be obscured in epoxy-embedded sections. All methods showed that hyphae possess an intricate filamentous matrix, which is apparently unoriented. However, the appearance of the cytoskeleton depended on the embedding medium, the solvent used during embedding, and whether cells were fixed by conventional fixation or freeze-substitution. PEG proved to be the best embedding medium for fungal cells because DGD caused cell shrinkage and produced a more lamellar than filamentous matrix. Also, the cytoskeletal filaments were clearer in freeze-substituted cells than in conventionally-fixed cells. Since we failed to find microtubules in the embedment-free sections, we re-embedded cells in Epon to discern whether microtubules or other cytoplasmic components had changed. Neither PEG nor DGD adequately preserved microtubules as compared to regular Epon-embedded sections, and other cellular structures were significantly altered. Therefore, alternative methods need to be employed to further characterize fungal cytoskeletons.     

Freeze-substitution and the preservation of diffusible ions

Freeze-substitution of biological material in pure acetone followed by low-temperature embedding in the Lowicryls K11M and HM23 yields stable preparations well suited for sectioning and subsequent morphological and microanalytical studies. Transmission electron microscopy of dry-cut sections shows that diffusible cellular thallium ions (Tl⁺) of Tl⁺-loaded muscle are localized at similar protein sites in freeze-substituted as in frozen-hydrated preparations. A comparison of X-ray micro-analytical data obtained from freeze-dried cryosections and sections of freeze-substituted normal (potassium-containing) muscle shows that K⁺ ion retention in the freeze-substituted sample is highly dependent on the freeze-substitution procedure used; so far, in the best case, about 67% of the cellular K⁺ is retained after freeze-substitution in pure acetone and low-temperature embedding. It is concluded that the retention of diffusible cellular ions is dependent on their interactions with cellular macromolecules during the preparative steps and that ion retention may be increased by further optimizing freeze-substitution and low-temperature embedding.     

The control of temperature during polymerization of Lowicryl K4M: there is a low-temperature embedding method

Methods are described for controlling the temperature of Lowicryl K4M in flat embeddings and in capsules during polymerization under ultra-violet (UV) irradiation in an Agar UVF 35 low-temperature cabinet. Aluminium blocks or an ethanol bath are used as heat ‘sinks’. Consequently the question posed by Ashford et al. (1986) — “Is there a low-temperature embedding method?” — is answered in the affirmative. This control of temperature is particularly important when using gelatin capsules since a temperature rise of as little as 2 K results in uneven polymerization.     

Ultrastructural localization of defined sequences of viral RNA and DNA by in situ hybridization of biotinylated DNA probes on sections of herpes simplex virus type 1 infected cells

The influence of fixation and enzymatic digestions on the ability of a denatured double-stranded DNA probe to bind specifically to related sequences of RNA and DNA in sections of Lowicryl embedded cells was investigated. Specificity of the hybridization was assessed using a biotinylated cloned subgenomic herpes simplex virus type 1 DNA fragment to localize viral nucleic acids in sections of infected cells. The probe was detected by anti-biotin antibodies and indirect immunogold labeling. Controls indicated that protease digestion of proteins from the section eliminated non-specific binding of the probe and labeling of endogenous biotin. Both formaldehyde and glutaraldehyde fixation retained viral RNA in protease digested sections. Its labeling was randomly and sparsely distributed over the fibrillo-granular network of the infected nucleus and over the ribosome-rich regions of cytoplasm. Labeling of single-stranded portions of viral DNA in protease-RNase digested sections was infrequent. It was located precisely over nucleoids of a few viral nucleocapsids whatever their location in the cell and their stage of maturation. Labeling of double-stranded viral DNA by denaturation of the DNA in the sections of Lowicryl embedded cells was possible after fixation with formaldehyde but not glutaraldehyde. Among several denaturation protocols, 0.5 N NaOH treatment was best for hybridization of both nonencapsidated and encapsidated viral DNA in protease-RNase digested sections. Free viral genomes were detected exclusively within the virus-replicating region of infected nuclei. Labeling of viral nucleoids was independent of their location in the cell. The high percentage of labeled viral nucleoids suggests that the related viral DNA sequence is not aggregated in the nucleoid but is extended and therefore numerous portions of this defined DNA sequence are accessible at the surface of the section for the binding of the probe.     

High-resolution low-temperature scanning electron microscopy for observing intracellular structures of quick frozen biological specimens

Intracellular structures of rapidly frozen biological tissues were observed in 3-D under a low-temperature scanning electron microscope using a newly developed side-entry type cryo-holder. The present low-temperature SEM is simple, easy to operate and effective for observing biological materials at high magnification. Biological tissues (the pancreas, small intestine, brown adipose tissue and Harderian gland) freshly removed from the mouse were immediately frozen in liquid propane cooled with liquid nitrogen, and their surfaces were manually fractured using a precooled razor blade in liquid nitrogen before introducing the cryo-holder into the SEM. When intracellular structures were revealed after appropriate sublimation, the specimens were coated with gold using a metal evaporator fitted to the side of the microscope column at one of the specimen chamber ports. The cryo-holder was connected to a copper braid coming from a liquid nitrogen reservoir to maintain a low temperature. Using this method, intracellular structures such as the mitochondria and endoplasmic reticulum were demonstrated at high magnifications. Ribosomal granules were discerned on the rough endoplasmic reticulum of the pancreatic acinar cells. Granular substances, presumably elementary particles, were also recognized on the mitochondrial cristae of the brown adipose tissue. The method was particularly effective for studying the 3-D configuration of lipid droplets which had been difficult to preserve by chemical fixation.     

Direct attachment of cell suspensions to high-pressure freezing specimen planchettes

We describe a procedure for high‐pressure freezing (HPF) of cultured cells using the HPF aluminium planchettes as a substrate. Cells are either grown directly on planchettes covered with Matrigel or allowed to attach to poly‐l ‐lysine‐coated planchettes. This method allows for rapid transfer of the cells into the HPF and minimizes physical and physiological trauma to the cells. Furthermore, the yield of well‐frozen cells approaches 100% for every cell type we have tried so far. In this report, we show well‐preserved ultrastructure in mitotic and interphase HeLa cells, isolated gastric parietal cells and isolated gastric glands. Immunogold labelling of H⁺/K⁺‐ATPase is shown in parietal cells of isolated gastric glands embedded in LR White resin. The aluminium planchettes appear to have little effect on cell physiology, as demonstrated by the fact that parietal cells cultured for 24–28 h on the planchettes retain their responsiveness to stimulation with histamine.     

Temperature control in Lowicryl K4M and glycol methacrylate during polymerization: is there a low-temperature embedding method?

An apparatus for embedding tissues at resin temperatures down to 228 K is described. By placing thermocouples in the resin the temperature has been monitored during embedding at low temperature with glycol methacrylate (GMA) and Lowicryl K4M. Even in this apparatus with a liquid cooling bath the heat of polymerization is not dissipated and the resin temperature rises. This rise is directly proportional to the resin temperature at the onset of polymerization and is higher in Lowicryl K4M than GMA. The initial resin temperature also affects the time taken for polymerization. The time to the onset of the peak and its duration are both increased as the temperature is lowered. This effect is more pronounced with GMA than Lowicryl K4M and polymerization of GMA is inhibited at the lowest temperature used. When Lowicryl K4M, polymerized at low temperature, is warmed up to ambient a further exothermic reaction occurs, which causes the resin temperature to rise well above ambient. Both this temperature peak and that during polymerization are reduced, but not totally eliminated, by reducing the resin volume. Aircooled systems are inefficient compared with the low-temperature apparatus used here and the resin temperature rise is consequently greater and, even with small resin volumes, it can be very high. It is therefore unlikely for published methods that the temperature specified has been maintained in the resin during polymerization. The implications of these findings are discussed in relation to enzyme and antigen survival. Recommendations include use of very small volumes of resin, refrigerated liquid-bath rather than air-cooled systems and contact with a heat sink when specimens are warmed up to ambient temperature. Examples of enzyme reaction, antigen survival and structural preservation obtained with the method are presented.     

Microtubule structure studied by quick freezing: Cryo-electron microscopy and freeze fracture

Microtubules have been quickly frozen and examined by electron microscopy using several techniques: (1) freezing of a thin layer of solution by plunging into cryogen, followed by cryo-electron microscopy of the unstained vitrified samples; (2) freezing by the propane-jet method, followed by freeze fracturing and metal replication. The unstained frozen-hydrated microtubules show a structure in agreement with X-ray diffraction data; they differ from negatively stained particles mainly by the better preservation of cylindrical shape. Secondly, they reveal a supertwist of the profilaments that is not detected reliably by other methods. This allows a determination of the number of protofilaments and the polarity. The structural resolution of unstained microtubules is similar to that of stained ones (about 2–3 nm); it is limited by low contrast and lack of crystalline order. Propane-jet or cryo-block freezing followed by freeze fracturing reveals the structures of the inner and outer surfaces of the microtubule wall at a resolution of 4 nm or better. The outside is dominated by the longitudinal protofilaments whereas on the inside one observes tilted cross-striations. Although the freezing temperatures of the two methods are different (liquid nitrogen or helium) they yield similar results for the case of thin layers of protein solution.     

Development of the gametophyte of the fern Schizaea pusilla

Schizaea pusilla is a pteridophyte with several unique developmental characteristics. In contrast to most other fern species, S. pusilla gametophytes remain filamentous throughout their development, and the gametophytes are associated with an endophytic fungus which appears to be mycorrhizal. In terms of tropistic responses, apical filament cells of young gametophytes are negatively phototropic compared with germ filaments of other ferns which exhibit positive phototropism. Cryofixation (propane jet freezing and high-pressure freezing) in conjunction with freeze substitution electron microscopy was used to study young gametophytes. The results demonstrate that apical filament cells have a distinctive structural polarity and that rhizoids also can be successfully frozen by these methods. The cytoskeleton and endomembrane system were particularly well preserved in cryofixed cells. In addition, Schizaea gametophytes were used as a test system to evaluate potential artifacts of propane jet freezing and high pressure freezing. There was little apparent difference in ultrastructure between cells cryofixed by either freezing method. These gametophytes will be useful in determining the effectiveness of cryofixation techniques and as a model system in tip growth studies.