|Home | About | Journals | Submit | Contact Us | Français|
Conventional heavy metal post staining methods on thin sections lend contrast but often cause contamination. To avoid this problem, we tested several en bloc staining techniques to contrast tissue in serial sections mounted on solid substrates for examination by Field Emission Scanning Electron Microscope (FESEM). Because FESEM section imaging requires that specimens have higher contrast and greater electrical conductivity than transmission electron microscope (TEM) samples, our technique utilizes osmium impregnation (OTO) to make the samples conductive while heavily staining membranes for segmentation studies. Combining this step with other classic heavy metal en bloc stains including uranyl acetate, lead aspartate, copper sulfate and lead citrate produced clean, highly contrasted TEM and SEM samples of insect, fish, and mammalian nervous system. This protocol takes 7–15 days to prepare resin embedded tissue, cut sections and produce serial section images.
For more than 50 years, thin section Transmission Electron Microscopy (TEM) has been a mainstay of cell biology. Early electron microscopists quickly realized the value of obtaining serial sections to gather three-dimensional (3D) information1. The application of serial section technique to the study of 3D microcircuitry of the nervous system began more than 30 years ago 2. New techniques have recently been developed to image larger volumes of tissue with the goal of understanding connectivity in the nervous system3. With the advent of serial block face (SBF-SEM) 4 and Focused Ion Beam (FIB) 5 methods for fully automated electron microscopic volume imaging and techniques for collecting long series of ultrathin sections onto large glass coverslip 6 or tape7 substrates, reflection mode Scanning Electron Microscopy (SEM) is moving to the forefront of electron microscopic 3D reconstruction. While traditional TEM methods have provided higher resolution and contrast imagery, they pose greater challenges for the collection and imaging of large specimens volumes due to the small grid sizes and the inherent fragility of the thin film supporting substrates required for transmission EM. Moreover, new Field-Emission SEM tools (FESEM) provide useful resolution that begins to approach that obtained by TEM, while the optimized en bloc staining methods we describe here provide contrast that also begins to rival that achieved by TEM.
The advantages of FIB- and SBF-SEM, which allow the collection of hundreds or thousands of automatically cut, perfectly aligned thin sections, over serial section TEM, which requires highly skilled collection and imaging techniques 8, is evident. However, these sections are lost permanently lost due to the destructive nature of the process of FIB and SBF-SEM. In contrast, the automatic collection of thin sections onto tape, and array tomography, where serial thin section ribbons are picked up onto carbon-coated coverslips (Fig. 1a, b), allow structural and multi-marker immunolabeling studies of the same section. Additionally, both tape and coverslip mounted samples permit repeated multi-site imaging of full sections without the interference of grid bars or unstable formvar films for the FESEM (Fig. 1c). Finally, the large chamber size of the FESEM easily accommodates serial sections mounted on these large specimen carriers i.e. 60 mm coverslips and 4-inch silicon wafers.
While major advances have been made in the field of imaging and cutting of sections, the same basic tissue preparation techniques, fixatives and heavy metal stains have not changed significantly in many years. We have combined and compared several classical EM preparation methods to achieve the best staining combination for segmentation of neuronal membranes in FESEM images. Here, we describe a method that can be applied to ultrastructural studies of many tissue types, both for the FESEM and TEM. For 3D serial reconstructions of different tissue types, one should consider which tissue elements are to be enhanced for imaging and choose the appropriate heavy metal stain to do so.
En bloc staining is applicable to both TEM and FESEM imaging, and it advantageous to be able to go directly to the electron microscope to view thin sections without the bother of further post staining. The main limitation of this method can be inadequate contrast due to uneven stain penetration and fragility of the tissue because of heavy metal infusion. Careful attention to solution preparation, incubation times and adequate rinsing steps will minimize background. Precise specimen handling at every step to lessen contamination and tissue loss is imperative when using the osmium impregnation technique.
The use of heavy metal on section staining methods using uranyl acetate and lead citrate 9,10 is widespread in classical Transmission Electron Microscopy (TEM) studies. The contrast enhancement observed in biological tissue is mainly due to the high affinity of uranyl and lead ions for proteins, nucleic acids, and hydroxyl groups in carbohydrates and RNA respectively 11. These positive stains, however, occasionally contaminate the sections with electron dense precipitates. Lead containing stains in particular, are notorious for causing fine pepper like precipitate that ruins the appearance of high-resolution images. In addition, the often laborious time spent staining and washing grids frequently damages formvar films and causes tears in the thin sections.
In 1960, Seligman and Hanker developed the OTO (osmium tetroxide-thiocarbohydrazide (TCH)-osmium) method12 which made use of thiocarbohydrazide as a bridging agent in the process called osmium impregnation. The osmiophilic TCH enhances the osmium staining of lipid components, and especially the cell membrane. TCH works by attaching itself to the osmium already present from the initial osmium fixation and acts as a bridge to allow deposition of additional osmium to the original osmium sites. A further advantage of using TCH is that the specimen is made more conductive to electrons by the additional metal staining. This helps prevent excess charging of the sample in the SEM13, 14,15.
The use of reduced osmium fixatives16,17 has also been shown to add contrast to specimens and preserve filaments and membranes. Addition of either potassium ferri- or ferrocyanide to the impregnation solution reduces the osmium (OsFeCN) causing it to be more reactive18. The reduced osmium stains membranes better and enhances labeling of sarcoplasmic reticulum and glycogen granules.
The use of imidazole, a tertiary amine, has been used to enhance the osmification of lipids, in particular those of cell membranes and unsaturated fatty acids19. In addition to the enhanced contrast of the plasma membrane, all intercellular membranes, including those of synaptic vesicles, the Golgi apparatus, endoplasmic reticulum, mitochondria and endosomes, have improved contrast20.
Many beautiful serial section TEM studies have provided 3D detail and relationships of structures in neuronal tissue, as shown by the studies of several labs21, 22,23,24,25. In these traditional serial section studies, sections are collected on 3 mm formvar coated slot grids. Successful collection requires excellent manual skills and dexterity and is often a major pitfall for the novice and expert alike. For example, sections can land on the copper grid edge, become folded or be affected by irregularities in the underlying support film26. In addition, formvar films are infamously hard to cast, extremely delicate and prone to instability and breakage in the electron beam. Another disadvantage is the size of the block face, which must be made extremely small to accommodate a great number of sections on a 3 mm grid. This is adequate for small areas of tissue but becomes problematic for large volume reconstruction of sizeable brain areas. Thus, there are several advantages to having serial sections on flexible tape or carbon-coated coverslips compared to formvar coated slot grids. The biggest advantage is the strength of the supporting material and ease of handling. For example, the use of the Jumbo Histo knife makes it relatively straightforward to pick up the ribbon of serial sections on a glass coverslip (Fig. 1b). The large boat size accommodates the coverslip, which is much easier to handle than numerous formvar coated grids. In addition, more and larger sections can fit onto a 20 x 60 mm coverslip (Fig. 1c) compared to a 3 mm formvar coated slot grid.
Our own search for improving contrast of embedded samples began with the development of SBF-SEM to map neuronal circuits, a tool that needed highly contrasted specimens for imaging. An en bloc protocol27 for SBF-SEM studies of the arterial wall was expanded and further developed in our labs specifically for staining neuronal tissue membranes and elements. We included the use of OTO because it had been used to make specimens conductive for SEM studies28.
In order to find the best staining protocol, we tested many permutations of different combinations of aqueous or buffered osmium, osmium imidazole, thiocarbohyrazide, carbohydrazide, copper lead en bloc, lead aspartate en bloc, potassium ferricyanide, and ethanolic phosphotungstenic acid (PTA). These were all tested in the presence or absence of uranyl acetate en bloc to determine the best combination and times to produce consistent, highly contrasted tissue. Many of variations of these staining techniques were developed over the past several years (Stains for tracing neural circuits by electron microscopy, J.B., Winfried Denk, Heinz Hortsmann, S.J S, Stuart Thompson and Ricardo Valenzuela, Society for Neurosci. Abstract, 2005; Staining for Electron Microscopic analysis of neural circuitry, J.B. and S.JS., Society for Neurosci. Abstract, 2007) and used most recently in SSEM (serial block face electron microscopy) studies of capillaries in the lung29 and connectivity in the retina30.
We studied various tissue samples that included mouse cerebral cortex, Drosophila central and peripheral nervous system and zebrafish Danio rerio optic tectum and retina. Fish and fly samples were microwave fixed and processed while mouse brain was perfusion fixed and either bench or microwave processed. Following sample embedment and microtomy, semi thin sections were examined for structural integrity (Fig. 2). Ultra thin sections (70 nm) were cut and were directly examined for evaluation by TEM with no post staining before preparing sections for the FESEM. Only those samples with enhanced contrast and good structure were further processed for FESEM. Serial ribbons of sections for FESEM studies were collected either on carbon-coated Kapton tape (DuPont) or carbon-coated, gelatin-coated coverslips and examined by FESEM. For one study for mouse cortex, ribbons on coverslips were imaged and the Reconstruct software program31 was used to generate a 3D image of an individual dually innervated synapse (Fig. 3).
Several different approaches can be taken to produce high contrast samples for electron microscopy. The combination of osmium, uranium and lead compounds are the basic heavy metal stains used for TEM studies, and one must test the different combinations and/or recipes to achieve optimum results for their individual preparations. It is important to consider the final goal and which cellular details need to be enhanced. In our case, we wished to have strongly stained neuronal membranes to aid in segmentation for 3D reconstructions from serial sections. This protocol can be adapted to work on a wide variety of tissue types for applications in several fields including cell biology, pathology, developmental biology and immunology.
Osmium tetroxide both stains and fixes tissue, while acting as a mordant to intensify other stains11. Reduced osmium fixatives produce higher contrast, particularly that of membranes and glycogen particles16. Reduced osmium fixatives have been reported as being variable due to the interaction of the chemical reaction with cellular components32 however we have not had this problem. Osmium imidazole (Osmid) also strongly contrasts membranes while leaving the cytoplasm electron lucent and highlighting microtubules. The downside of using imidazole compounds however is that they have a tendency to cause osmium to precipitate out of solution if the pH is not carefully controlled. One must evaluate the ultrastructural details to be reconstructed or segmented when choosing which osmium fixative to use.
Uranyl acetate (UA) is commonly used in conventional TEM tissue preparation, primarily as a section stain but also en bloc during processing of wet tissue. It is relatively non-specific stain, primarily staining proteins. It can be less effective when used en bloc as compared to section stain and can react with the buffers used during specimen processing. Often, the addition of ethanol or methanol to the staining solution improves penetration into the tissue. Using uranyl acetate in combination with lead stains has an additive effect since UA acts as a mordant for deposition of more stain33. We recommend that the researcher evaluate the sample with and without aqueous UA en bloc stain to determine which method produces acceptable contrast.
Lead compounds are routinely used for TEM studies, and stain a variety of cellular components including membranes, nucleic acids and glycogen34. The main drawback of lead containing stains is the propensity for lead to form insoluble lead carbonate precipitates upon exposure to carbon dioxide. The use of en bloc lead stains avoids deposition of lead carbonate precipitate that forms during section staining on the bench, while boosting electron density of the sample35. The combination of lead citrate and copper sulfate used together en bloc provide definitively stained membranes while leaving the cytoplasm electron lucent. According to Hyatt33 lead aspartate en bloc stain penetrates more slowly than lead citrate and ultrastructural details may be compromised. In our hands, we found that lead aspartate en bloc did not affect the quality of the sample (see Fig. 4). On the contrary, it has been reported that the combination of potassium ferricyanide reduced osmium and lead aspartate en bloc lead produce the best high contrast results for electron microscopic radioautography36.
We have used osmium impregnation coupled with either buffered imidazole or potassium ferricyanide reduced osmium and found that it provides excellent membrane contrast for both TEM and SEM studies. It also makes the sample conductive by increasing the production of secondary electrons thus avoiding sample charging.
Our method avoids the loss of material at the staining stage, provides clean and evenly stained serial sections while saving time. Cellular details including synaptic vesicles, microtubles, endoplasmic reticulum, post-synaptic densities, Golgi apparatus and cell membranes are easily discernible in the FESEM (Fig. 5). Additionally, microwave fixation results in improved structure, especially for hard to preserve specimens like zebrafish brain (Fig. 6) (J.B., Meyer, M., Niell, C. & S.J S. Mol. Biol. Cell (suppl), abstract 672, 2003) and Drosophila37. This protocol has been found to work well for a wide variety of neuronal tissues including mouse brain, fruit fly Drosophila (Fig. 7) and zebrafish Danio rerio both for FESEM and TEM studies (Supplementary Figures 1 and 2).
Microwave energy has been used as a tissue preparation method for light and electron microscopy for many years38, 39,40,41. Microwave energy improves fixation quality, speeds up processing times considerably and enhances antigenicity and immunolabeling 42,43. Conflicting ideas about the nature of microwave fixation remain on the mechanisms of the heating versus the microwave effects. The heating effect is believed to improve penetration of reagents and speed up fixation44. Other publications have suggested that the microwave effects were non-thermal based on a study of globular proteins45 and an increased rate of bone demineralization independent of temperature46. We agree with the finding that the microwave effect itself results in improved structure and faster processing times47. Thus newer model laboratory microwave ovens (i.e. Ted Pella Biowave®) are equipped with load coolers and wattage controllers to carefully control the temperature of the specimen during the processing steps. In this protocol, the primary preparation steps are done using a commercially available specially outfitted microwave oven. It is recommended that when used in combination with osmium impregnation, one should use bench resin infiltration instead of microwave resin infiltration to insure adequate penetration of the resin into the heavily osmicated tissue. All other steps, including fixation and dehydration, can be done using the microwave oven or bench method if a special laboratory microwave is not available.
Zebrafish and Drosophila preparations were processed with a Ted Pella microwave oven equipped with a wattage controller, water re-circulator and chiller. Mice were perfused through the heart with glutaraldehyde and paraformaldehye mixture using either a Perfusion One or Perfusion Two system (Leica). After the brain was removed from the animal, it was sectioned with a vibratome into 100–200 μm thick sections and further processed for electron microscopy using either microwave or bench-processing methods as described in the protocol.
The initial aldehyde fixation is very important for optimizing good ultrastructure. It is important to work quickly and expose the tissue to the fixation solution, especially during perfusion fixation (see step B iii). If dissecting tissue, work quickly and immerse in fixative rapidly. We have found microwave fixation works best for zebrafish and Drosophila samples, and fast and careful perfusion works best for mammalian brain. Zebrafish are especially difficult to get high quality ultrastructure of the nervous system, perhaps because of the water content of the tissue. The health of the animals and precise attention to detail is important when preparing and running up the tissue for EM studies, whether SEM or TEM. Use only freshly prepared reagents and healthy animals in insure the best outcome and try several batches of animals or cells. Often times, it takes several attempts to achieve high quality results.
While osmium impregnation with OTO makes the tissue more electron dense and conductive, it also makes the tissue more fragile and difficult to infiltrate especially if the volume is large. It is important to find just the right amount of osmium deposition, to rinse well with distilled water between steps and to dehydrate and infiltrate the tissue thoroughly. Specimens should be carefully handled during processing by using fine brushes to avoid tissue breakage.
During processing, freshly opened bottles of propylene oxide should be used for final dehydration steps and fresh epon mixtures used for embedding because of possible water contamination. Embedded tissue should be hardened thoroughly for two days before attempting to cut the blocks. When thin sectioning the blocks, the trimming step is critical. It is imperative that the sides be evenly angled on the trapezoid and that the top and bottom of the block are parallel in order to obtain a straight ribbon. Either a diamond trimming knife or freshly opened razor blade and steady hands can be used.
Finally, it is important to do appropriate controls if the results are uneven- checking the pH and osmolarity of the fixative, making fresh solutions, omitting the OTO or lead en bloc steps, and trying different time points. Some trial and error is necessary to adapt this protocol to different tissue types. Our protocol is presented as what works best for our test material, but different permutations in the choice of fixatives and resin mixtures can be employed. We recommend reviewing the types of stains available for all EM studies and highly recommend the book “Stains and Cytochemical Methods” for reference.
Perfusion and immersion fixation
Cover slip subbing solution Dissolve 1.5 g of gelatin in 290 ml of distilled water by heating to 60°C. Dissolve 0.15g of chromium potassium sulfate in 10 ml of dd water. When gelatin solution cools down to 37°C, combine the two solutions, filter and pour into staining tank. Use freshly prepared solutions only. Dip coverslips in rack into solution. Allow to dry overnight in dust free environment.
Tricane (Tricaine methanesulfonate, TMS, MS-222 Sigma Cat. No. A-5040). Add 40 mg to 10 mls dd water. Make fresh right before use.
Cacodylate Buffer purchased from EMS 0.2M pH 7.4. Ready to use. Store at 4°C.
Aldehyde Fixatives Use ampoule cracker to open vials of fixative. For perfusion fixation add 20 mls of 32% wt/vol paraformaldehyde to 80 mls 2X PBS. Add 8 mls of 50% (vol/vol) glutaraldehyde and 52 mls of dd water. Make fresh before using. For insects and fish microwave immersion fixation, use 2% vol/vol glutaraldehyde and 2 % paraformaldehyde in O.1M cacodylate buffer pH 7.4. Mix 4 mls of 8% vol/vol glutaraldehye with 2 mls of 16% wt/vol paraformaldehye in 8 mls of 0.2M cacodylate buffer. Add 2 mls of dd water to make final volume of 16 mls. Make fresh right before use.
Osmium tetroxide 2% wt/vol in 0.1 M cacodylate buffer. Use ampoule cracker to open vials of fixative. Add equal amounts of 4% wt/vol aqueous osmium tetroxide and 0.2M cacodylate buffer pH 7.4 in a 20 ml scintillation vial. This fixative can be used for 1 or 2 days. Discard if the solution looks blackish (oxidized). Store at 4 °C inside of a secondary glass jar.
Phosphate buffered saline (PBS) 2X(0.02M pH 7.4) mix 1 pouch of powder with 500 ml of dd water. Can be stored for several weeks at 4 °C. If debris appears in bottle, discard.
Thiocarbohydrazide (TCH) 1% wt/vol aqueous solution. Heat 25 mls of dd water on a hot plate to 58°C. Add 0.25 g of thiocarbohydrazide to the heated water while stirring until powder is dissolved, can leave 2–3 hours, color will darken. Turn down heat while stirring. Remove from hotplate and let cool to room temperature (25 °C). Filter using a 0.45 μm syringe filter. Store at room temperature to avoid precipitation for up to 1 week.
Potassium Ferricyanide 3% wt/vol. Add 1.5 g of powder to 50 mls of 0.2M sodium cacodylate buffer, pH 7.4. Shake until dissolved. Bright yellow solution. Store at 4° C, good for several months.
Imidazole Add 0.7 g to 50 mls of sodium cacodylate buffer pH 7.4 and stir until dissolved. Adjust pH to 7.4. Store at 4° for weeks. ▲ CRITICAL STEP Be certain to pH the buffer solution to 7.4 after adding the imidazole or it will precipitate out when the osmium tetroxide is added.
Osmium/Imidazole (Osmid) Add equal amounts of 4% aqueous osmium tetroxide and imidazole/cacodylate solution. ▲ CRITICAL STEP Must adjust the pH of the osmium/imidazole solution or the osmium will precipitate out of solution. Use pH paper and add 1 N sodium hydroxide to raise the pH to ~7.4. Amber color is normal. Don’t use if precipitate is evident. Make fresh right before use for all osmium steps.
Reduced Osmium Fixative(RO) Open fixative vial using ampoule cracker. Add equal amounts of 4% wt/vol aqueous osmium tetroxide to 3% potassium ferricyanide in 0.2M cacodylate buffer. Store in double jar in refrigerator at 4 °C for 1–2 days.
First make up four components as follows. For 1M lead nitrate, dissolve 3.31 g in 10 mls of dd water on a stirrer until dissolved. For 0.4M copper sulfate dissolve 0.64 g in 10 mls of dd water on a stirrer until dissolved. For 1M sodium hydroxide dissolve 4 g of sodium hydroxide pellets in 40 mls of dd water. For 0.2 M trisodium citrate mix 0.6 g in 10 mls of dd water on a stirrer until dissolved.
Then start with a small beaker containing 1.75 mls of dd water and 3.25 mls of sodium citrate with constant stirring. Add 0.5 mls of 1M lead nitrate and 0.125 mls of copper sulfate followed by 1 ml of 1 M sodium hydroxide (carbonate free) until blue solution clears. The blue solution should be precipitate free and used immediately. Spin down 12,578 g for 5 mins. Then filter with a 20 μm syringe filter before using. Make fresh each time from stock solutions. Keep stock solutions at 4° C for weeks.
Make up aspartic acid stock by dissolving 0.998 g of L-aspartic acid (Sigma Cat. No. A4534) into 250 ml dd water. It dissolves better at pH 3.8. Store for up to 1 month. To make heavy metal stain, dissolve 0.066 g lead nitrate in 10 ml of the aspartic acid stock; adjust pH to 5.5 with 1M KOH. Heat the solution to 60° C before use to ensure stain stability. Use only freshly prepared staining solution.
Mix 19 mls of epoxy resin with 12.5 mls of DDSA and 10 mls of NMA in a tripour beaker on a small stir plate in the fume hood. Stir components together for 1 hour. Add 0.7 mls of BDMA under the surface and continue stirring for another 15–30 mins. Store unused portions in 15 ml tubes in the freezer at −25 °C for infiltration. Use only freshly prepared mixtures for final embedding.
If using fly, using a brush or fine tipped pipet, carefully transfer fly heads to small vials containing distilled water or 50% ethanol. For fish and brain slices leave in glass vials used for fixation unless the old ones appear contaminated with previous solutions then transfer samples to fresh ones containing a solution of dd water or 50% ethanol using a fine brush.
▲ CRITICAL STEP Work quickly and do not let tissue dry out. Leave a small amount of ethanol in the vial and do not leave tissue uncovered. ! CAUTION tissue is brittle handle with care!
See Table 1 for troubleshooting guidance.
The assessment of well-prepared material takes several steps. When trimming the block for initial cutting, the hardness and infiltration of the embedded tissue should be evident. If using a razor blade to trim, the resin should feel hard and trim easily, not rubbery. When trimming the block with the Leica EM trimmer, the embedded tissue should look shiny after a smooth pass with the trimming tool, and there should be no big chips present in the specimen face.
Semi thin sections should have a dark brown color resulting from the impregnation of osmium, as in Fig. 2 and the structure should be visible by LM even without toluidine blue staining. Gross morphological aberrations such as empty spaces and white holes are indicative of poor tissue preparation-possibly fixation, dehydration or infiltration. For example, if chunks fall out of the section or block or the block face looks shredded or chipped, this indicates poor infiltration. While it is impossible to fully evaluate the quality of fixation at the LM level, it is possible to recognize poorly processed material. It is not worth it to spend any more time with poorly processed samples. The best specimens have great contrast at the LM and EM levels, sharp looking membranes and overall integrity of the structure. See Supplementary Figure 3 for a comparison of en bloc staining methods for more detail.
Thin sections should cut well and have good contrast in the both FESEM and TEM without any section staining. Contrast will be reduced in the SEM compared to the TEM, but a kV of 7–8, working distance of 6–9 mm, contrast and brightness settings of 50–70 range and a slow scan speed 8–10s with a dwell time of 1 μsec./pixel should provide a good image without any specimen damage. These conditions work well in our imaging system but may differ for different instruments and detectors.
It is important to let the sample air dry or on a slide warmer then in a vacuum desiccator after picking up the sections on coverslips, as moist samples will cause “bubbling” of sections in the FESEM and contamination of the apertures.
|1C||Brain is soft rather than hard||Poor perfusion||Discard, try another animal|
|Brain appears bloodied||Discard, try another animal|
|Liver does not turn pink||Discard, try another animal|
|Fixation solution coming out of the nose or mouth of animal||Discard, try another animal||Use ideal perfusion pressure (~150 mmHg)|
|1, 2, 43||Tissue dark and murky looking, uneven membranes||Poor initial fixation, unhealthy tissue||Use only healthiest tissue for experiments, If poor perfusion or sick animals, discard. Use fresh fixatives only|
|Inadequate rinsing between osmication and TCH treatment||Rinse well before and after osmium and TCH steps.|
|23, 25||Block chipping or tissue falling out of section||Inadequate infiltration||Be sure to infiltrate overnight on rotator for 2 nights|
|Using EM trimmer blade too fast||Approach block slowly|
|Block center soft||Inadequate infiltration||Longer infiltration, use only freshly prepared resins for infiltration and embedding|
|28.32||Ribbon curves||Top and bottom of block face not parallel, uneven sides||Re-trim block with fresh razor blade or diamond trimming tool|
|29||Ribbon not sticking together||Not enough glue on top and bottom of block face||Add more glue to top and bottom of block face|
|44||Charging of specimen in SEM||Not enough carbon coating or copper tape on coverslips||Remove sample and try more tape or carbon coat the surface|
|Beam damage, breaks in section||Beam too bright||Try using lower voltage, reduce kV and brightness|
|Bubbling of sections, black blobs forming||Sections have too much moisture||Allow sections to air dry then store under vacuum for 24 hours|
We thank the following foundations, companies and support funding for this work: Gatsby Charitable Trust, John S McDonnell Foundation, McKnight Foundation, The Mathers Foundation, Center for Brain Science, Harvard, Initiative for Innovative Computing, Microsoft Research, Carl Zeiss SMT, JEOL and FIBICS.
We appreciate helpful discussions with Winfred Denk, John Heuser, Tom Reese and Morris J. Karnovsky. We acknowledge Nafisa Ghori and Erika Hartwig for technical support, Kristina Micheva and Georgeanne O’Brien for comments on the manuscript, Cristel Genoud and Joel Mancuso from Gatan for SBF-SEM imaging of zebrafish samples, and Rick Giberson of Ted Pella Inc. for advice on microwave conditions.
The authors of this manuscript declare that they have no competing financial interests.
Author contributionsJB developed the staining concept, prepared the fish and Drosophila samples and prepared the manuscript.
JCT, NK, and RS imaged the Drosophila sections.
RS assisted with the sectioning, imaging and overall block quality assessment.
KH, RS, JCT and NK improved ultra thin sectioning and collection.
KH developed the method of collecting ultrathin sections on tape and built the ATUM devices used.
JWL helped motivate the effort to find better en bloc staining protocols, oversaw all the imaging experiments that were carried out in his laboratory and helped interpret the image data.
SJS helped motivate the effort to improve en bloc staining, and oversaw and assisted with imaging experiments carried out at Stanford.