Dual effect of tannic acid on the preservation and ultrastructure of phosphatidyl choline vesicles

Summary The use of tannic acid has been proposed to improve the preservation of phospholipids in tissues. We investigated the effects of tannic acid on the preservation of small unilamellar vesicles, prepared from sonicated aqueous suspensions of phospholipids.With cryo-electron microscopy it is demonstrated that small unilamellar vesicles are formed after sonication of the phospholipid suspensions. Fixation of vesicles without tannic acid results in extraction of the phospholipids during dehydration and embedding. Fixation of vesicles containing phosphatidyl choline with tannic acid, with or without glutaraldehyde, results in a fast (within a second) aggregation of the vesicles and the resulting sediment can be dehydrated and embedded when a postfixation in osmium tetroxide is carried out. Small unilamellar vesicles fixed in this way are retrieved in thin sections as multilamellar vesicles with a periodicity of about 5 nm for dimyristoylphosphatidyl choline and about 6 nm for dioleoylphosphatidyl choline.By using ¹³C-phosphatidyl choline it was also demonstrated that tannic acid prevents to a large extend the extraction of phosphatidyl choline during fixation, dehydration and embedding. This dual effect of tannic acid on phosphatidyl choline, aggregation and fixation, should be considered when using tannic acid in tissue preparation.     

Simple and rapid methods for SEM observation and TEM immunolabeling of rubber particles.

We developed a method involving air-drying of a rubber suspension after fixation in glutaraldehyde-tannic acid and postfixation in osmium tetroxide for SEM observation. For TEM immunolabeling the suspension was air-dried after osmium-only fixation. Whereas conventional methods failed to satisfactorily stabilize rubber particles, the methods described here proved successful in preserving their integrity.     

Vapor fixation for immunocytochemistry and X-ray microanalysis on cryoultramicrotome sections

Several organic and inorganic vapor fixatives have been tested for their ability to stabilize the ultrastructure of freeze-dried thin cryosections. The vapors from osmium tetroxide and dry formaldehyde gave a good preservation of the ultrastructure. Fixation in formaldehyde vapor preserved the immunoreactivity of alpha-amylase in exocrine pancreas, as was demonstrated with an indirect labeling technique using anti-alpha-amylase and protein A-gold. A major advantage of the use of vapor fixation is that cryosections from a specimen of fresh-frozen tissue can be used for immunocytochemistry as well as for X-ray microanalysis, as was demonstrated for the exocrine pancreas. This opens the possibility of localizing atomic species (X-ray microanalysis) and molecular species (immunocytochemistry) at the subcellular level on thin cryosections from the same tissue block.     

PREPARATION OF ISOLATED RAT LIVER MITOCHONDRIA FOR ELECTRON MICROSCOPY

A comparative study of the fixation of isolated rat liver mitochondria was undertaken. If the criterion is adopted that after processing, the mitochondria should resemble as closely as possible rat liver mitochondria in situ, the procedure found to produce such preservation was that of fixation in suspension in veronal-buffered 2% potassium permanganate. Fixation in osmium tetroxide produced variable results, while mitochondria fixed in glutaraldehyde were contracted. We suggest that in cases where fixation procedures modify the morphological appearance of mitochondria, the significance of such changes must be treated with caution.     

Effect of chemical fixatives on accurate preservation of Escherichia coli and Bacillus subtilis structure in cells prepared by freeze-substitution.

Five chemical fixatives were evaluated for their ability to accurately preserve bacterial ultrastructure during freeze-substitution of select Escherichia coli and Bacillus subtilis strains. Radioisotopes were specifically incorporated into the peptidoglycan, lipopolysaccharide, and nucleic acids of E. coli SFK11 and W7 and into the peptidoglycan and RNA of B. subtilis 168 and W23. The ease of extraction of radiolabels, as assessed by liquid scintillation counting during all stages of processing for freeze-substitution, was used as an indicator of cell structural integrity and retention of cellular chemical composition. Subsequent visual examination by electron microscopy was used to confirm ultrastructural conformation. The fixatives used were: 2% (wt/vol) osmium tetroxide and 2% (wt/vol) uranyl acetate; 2% (vol/vol) glutaraldehyde and 2% (wt/vol) uranyl acetate; 2% (vol/vol) acrolein and 2% (wt/vol) uranyl acetate; 2% (wt/vol) gallic acid; and 2% (wt/vol) uranyl acetate. All fixatives were prepared in a substitution solvent of anhydrous acetone. Extraction of cellular constituents depended on the chemical fixative used. A combination of 2% osmium tetroxide-2% uranyl acetate or 2% gallic acid alone resulted in optimum fixation as ascertained by least extraction of radiolabels. In both gram-positive and gram-negative organisms, high levels of radiolabel were detected in the processing fluids in which 2% acrolein-2% uranyl acetate, 2% glutaraldehyde-2% uranyl acetate, or 2% uranyl acetate alone were used as fixatives. Ultrastructural variations were observed in cells freeze-substituted in the presence of different chemical fixatives. We recommend the use of osmium tetroxide and uranyl acetate in acetone for routine freeze-substitution of eubacteria, while gallic acid is recommended for use when microanalytical processing necessitates the omission of osmium.     

Lipid fixation during preparation of chloroplasts for electron microscopy

Reaction of osmium tetroxide with isolated spinach chloroplasts fixed completely the glycolipids, phosphatidyl glycerol, and phosphatidyl choline. Under the same reaction conditions only 30% of the chlorophyll was fixed. Reaction of potassium permanganate with isolated spinach chloroplasts fixed more than 90% of the glycolipids, phosphatidyl glycerol, and phosphatidyl choline, provided the reaction period was long enough. Potassium permanganate also fixed the chlorophyll. Reaction of osmium tetroxide and potassium permanganate with isolated (14)C-lipids from Chlorella pyrenoidosa fixed 59% and 66% of the radioactivity, respectively. The lipids that were not fixed included sterols and pigments. Electron micrographs show that chloroplasts extracted with chloroform-methanol after fixation in osmium tetroxide or potassium permanganate differ from those dehydrated with acetone mainly in that in the former, osmiophilic globules have been removed and there seems to be some fusion of the boundary membranes and grana membranes. These effects may be due to the extraction of unfixed, neutral lipids such as sterols and quinones.     

A fixation procedure suitable for autoradiography of endogenous long-chain acyl carnitine

A tissue processing procedure was evaluated for fixation of endogenous long-chain acyl carnitine (LCA) to facilitate autoradiographic subcellular localization of this amphiphile. Suspensions of neonatal rat myocytes labeled with exogenous 14C-palmitoyl carnitine retained 85.2% of the radiolabel after tissue processing. Autoradiography demonstrated no significant translocation of radiolabeled LCA from myocytes to unlabeled sheep erythrocytes mixed in equal proportions and processed together. To evaluate endogenous LCA fixation, cultured myocytes were incubated for 3 days with 3H-carnitine. Radioactivity was distributed in LCA, short-chain acyl carnitine, and free carnitine pools in proportion to the physiological concentrations of the metabolites traced. Before tissue processing, LCA contained 4.5% of total radioactivity. After tissue processing, labeled water-soluble components were lost and 88% of the retained radioactivity was in the LCA pool. The enrichment of endogenous LCA radioactivity was attributable to the selective extraction of endogenous short-chain and free carnitine. Nearly 75% of endogenous LCA was preserved. In contrast, 99.5% of both endogenous short-chain and free carnitine were extracted. Thus, endogenous LCA can be selectively preserved, permitting quantitative subcellular localization of this amphiphile with ultrastructural autoradiography.     

Preservation of Tracheal Mucus by Nonaqueous Fixative

Two nonaqueous fixatives, composed of fluorocarbon solvents with dissolved osmium tetroxide, were used to determine the feasibility of preserving the mucous coat in bovine and rat trachea for light and electron microscopy. Aqueous fixatives, while providing excellent cytological preservation, wash away the mucous lining, precluding ultrastructural analysis. Inclusion of ruthenium red or alcian blue within aqueous fixative improved retention of mucus, but provided incomplete, patchy results. Fixation with nonaqueous fluorocarbon solvent and dissolved osmium tetroxide preserved a continuous mucous epiphase layer above a clear hypophase layer. Subcomponents of the mucus included an electron dense surface layer, interrupted patches of mucus above the surface layer and electron dense membrane-like material within the mucus. This method of fixation will preserve mucus for light, scanning and transmission electron microscopy, using either intratracheal or immersion methods of fixation. The latter would enable use of materials from large animal models, autopsy or an abattoir.     

Influence of fixation procedures on the microanalysis of lead-induced intranuclear inclusions in rat kidney

Using Laser Microprobe Mass Analysis (LAMMA), we studied the chemical composition of lead-induced intranuclear inclusions in rat kidney tissue prepared by three different wet chemical fixation procedures for transmission electron microscopy. Fixation with glutaraldehyde-Na2S gave the same results as fixation with glutaraldehyde only: a high lead concentration could be detected. Therefore, for lead strongly bound to proteins, precipitation procedures are not essential. Post-fixation with osmium tetroxide drastically changed the composition of the inclusions: the lead concentration decreased substantially, while sodium, calcium, and barium were introduced. The osmium tetroxide fixative was found to be the source of the contamination. It also contained aluminum, and we suggest that other proteins (e.g., in neurofibrillary tangles) might be able to take up Al out of solution and that care must be exercised in interpreting the microanalytical results of osmium-fixed material. For the microanalysis of the lead inclusions, fixation with glutaraldehyde only provides a good compromise between preservation of the ultrastructure and maintenance of the element distribution.     

Rapid preparation of uncoated biological specimens for scanning electron microscopy.

We have developed a relatively rapid glutaraldehyde-tannic acid (GTA) and osmium tetroxide (OsO4) fixation procedure which permits many types of uncoated biological specimens to be examined in the scanning electron microscope (SEM) at 20 kV without the occurrence of charging. Most specimens taken one day can be examined in the SEM the following afternoon. Types of specimens successfully treated were perfused adult and embryonic rat tissues, confluent human skin fibroblast tissue cultures, plant roots, flowers, seeds, some garden insects, and microcolonies of salivary streptococci. Cells in suspension and extracted human teeth did become electron conductive when treated with the GTA procedure. Most suspended cells must be centrifuged between each solution and the GTA procedure increases the preparation time for these cells. Extracted teeth are usually simply dried and coated. Therefore, the usual SEM preparation techniques are shorter and perhaps more useful for these types of specimens.     

Preparation of tissues for localization of sex hormones by autoradiography

Summary Tissues of rats given ³H-oestradiol were prepared for autoradiography according to methods commonly used in light and electron microscopy.By formalin fixation large amounts of radioactive material were lost, both in the fixative and during dehydration. Altogether 78.6±7.5 per cent was extracted from uterine tissue, while 49.0±4.6 per cent was lost from liver tissue removed 15 minutes after the injection. Significantly more radioactivity was lost in the fixative from liver tissue than from uterine. In the former fixation accounted for about 60 per cent of the loss, whereas in the latter it was responsible for about 25 per cent.Osmium tetroxide fixation was found to retain the radioactivity of liver and uterine tissue almost completely. However, large amounts were invariably extracted during dehydration. Although only 3.9±1.2 per cent of the radioactivity of uterine tissue diffused into the fixative, 72.8±12.4 per cent was extracted during ethanol dehydration. A heavy loss was also registered when dehydration and infiltration were carried out in glycol methacrylate.Glutaraldehyde perfusion and postfixation with osmium tetroxide retained almost completely the radioactivity of uterine and pituitary tissue. Nevertheless, nearly all of it was extracted during ethanol/propylene oxide dehydration and Epon embedding.The methods studied are not adequate for accurate autoradiographic localization of oestradiol.This work was supported by grants from The Norwegian Cancer Society and by Nordisk Insulinfond. The skilful assistance of Miss Helga Friedl and Mrs. Jane Larsen is gratefully acknowledged.     

Improved Fixation of Cellulose-Acetate Reverse-Osmosis Membrane for Scanning Electron Microscopy

Fixation of cellulose-acetate membranes with either glutaraldehyde-osmium tetroxide or glutaraldehyde-ruthenium tetroxide resulted in extensive electron beam damage. Beam damage was eliminated and the bacterial surface structure was preserved, however, when cellulose-acetate membranes were fixed with glutaraldehyderuthenium tetroxide and treated successively with thiocarbohydrazide and osmium tetroxide.     

Immunoelectron microscopic visualization of neurohypophyseal hormones: evaluation of some tissue preparations and staining procedures.

To obtain optimal electron microscopical localization of vasopressin in the rat neurohypophyses two immunocytochemical staining procedures and several tissue treatments were evaluated. The peroxidase-antiperoxidase technique allowed greater dilution of the first antibody than the indirect method using a commercial peroxidase conjugate. This appeared crucial for the dilution-dependent specificity in the localization of vasopressin. Immersion fixation with formalin gave better results than those obtained with perfusion fixation with glutaraldehyde-paraformaldehyde (resulting in similar preservation of immunoreactivity) and freeze substitution (showing the best preservation of immunoreactivity). However, these latter two tissue fixation methods resulted in less than optimal preservation of general ultrastructure than immersion fixation in formalin alone. Immersion fixation with glutaraldehyde-paraformaldehyde followed by OsO4 improved ultrastructural detail, but immunoreactivity decreased considerably. Fixation with paraformaldehyde-picric acid resulted in poorest preservation of morphologic detail. Immunoreactivity was similar in both Epon 812 and Araldite 6005 embedded tissue.     

Swelling of golgi apparatus cisternae in monensin-treated tissues is modulated by fixation conditions

Summary Swelling of Golgi apparatus cisternae is reported to be a common response to the ionophore, monensin. However, the amount of swelling depends on fixation, thus raising the question of whether the swelling response is due to monensin or to the fixation protocol. To resolve this problem, maize root cap cells were treated with monensin and then fixed with glutaraldehyde and osmium tetroxide (applied sequentially), osmium tetroxide alone, or aqueous potassium permanganate, or were quick frozen in liquid propane and substituted in acetone-osmium tetroxide. The chemical fixatives (which take minutes to stabilize tissue elements) were judged by comparison with freeze substitution which requires only fractions of a second to stabilize tissue elements. The results verify that monensin causes cisternal swelling and that this swelling is best observed at the ultrastructural level by fixation in glutaraldehyde/osmium tetroxide or by freeze substitution.     

A Comparison of Standard Lipid Staining Techniques Used in Electron Microscopic Studies of Mammalian Tissues

Comparisons of several standard techniques for staining lipids in ultrastructural studies have been undertaken using the rat uterine epithelium as the experimental tissue. The best technique for clarity, retention of stain, and acceptability of cellular ultrastructure utilized p-phenylenediamine after primary fixation in glutaraldehyde and postfixation in osmium tetroxide. While osmium by itself stained only unsaturated lipids and p-phenylenediamine stained no lipids in spot tests, when acting together, the staining of unsaturated lipids was enhanced and some staining of saturated lipids was seen. Further, the marked extraction of stained lipids normally found during dehydration did not then occur.     

A comparison of standard lipid staining techniques used in electron microscopic studies of mammalian tissues

Comparisons of several standard techniques for staining lipids in ultrastructural studies have been undertaken using the rat uterine epithelium as the experimental tissue. The best technique for clarity, retention of stain, and acceptability of cellular ultrastructure utilized p-phenylenediamine after primary fixation in glutaraldehyde and postfixation in osmium tetroxide. While osmium by itself stained only unsaturated lipids and p-phenylene-diamine stained no lipids in spot tests, when acting together, the staining of unsaturated lipids was enhanced and some staining of saturated lipids was seen. Further, the marked extraction of stained lipids normally found during dehydration did not then occur.     

Structure of the nucleoid in cells of Streptococcus faecalis

The structure of the nucleoid of Streptococcus faecalis (ATCC 9790) was examined and compared in the unfixed and fixed states by immersive refractometry and electron microscopy. It appears from these studies that the nucleoid structure is much more centralized in unfixed chloramphenicol-treated (stationary-phase) cells than it is in cells in the exponential phase of growth. The more dispersed configuration of the exponential-phase nucleoid could be preserved by fixation in glutaraldehyde, but not in Formalin or in osmium tetroxide. One important factor in explaining these differences in preservation is that glutaraldehyde (but not Formalin or osmium tetroxide) can rapidly cross-link the amino groups of macromolecules in cells. It was also observed that osmium tetroxide resulted in a preferential breakdown of nascent ribonucleic acid. These results are interpreted as indicating that glutaraldehyde is able to stabilize the exponential-phase nucleoid before it assumes the more central appearance seen in osmium tetroxide- and Formalin-fixed cells. These results are discussed in terms of the proposed organization of the exponential-phase nucleoid in unfixed cells.     

A study of the ultrastructural organization of cytochromec-phospholipid membranes as revealed by various experimental treatments

Summary Two different types of cytochromec-phospholipid model membranes have been investigated by electron-microscopy. The preservation of the lamellar structure of the cytochromec plus phospholipid-water system after extraction of 88% phospholipids with aqueous acetone followed by only aldehyde fixation is revealed.Pronase digestion of this type of model membrane leads to removal of 80% of the protein and loss of ordered structures as revealed by aldehyde post-fixation.It is shown that the hydrated isooctane-soluble cytochromec-phospholipid complex retained the lamellar structures after extraction of 75% phospholipids followed by separate aldehyde and osmium tetroxide fixations. Pronase treatment of this type of model membrane leads to removal of 25 to 30% of the protein. After aldehyde fixation of these pronase-treated samples the triple-layered structures are preserved.The structural integrity of both types of model membranes after organic extractions with aldehyde pre- or post-fixation is thought to be the result of cross-linkage of the protein by aldehyde polymers on both surfaces and through lipid bilayers. Possible localization of cytochromec in investigated model membranes is discussed.     

HYDROXYPROPYL METHACRYLATE, A NEW WATER-MISCIBLE EMBEDDING MEDIUM FOR ELECTRON MICROSCOPY

Aldehyde-fixed rat tissues were variously dehydrated and impregnated in water-miscible 2-hydroxypropyl methacrylate (HPMA) containing 3 to 20 per cent water and 0.1 per cent α,α-azobisisobutyronitrile as catalyst for subsequent polymerization with ultraviolet light. Heat polymerization was also effective. Blocks of embedded tissue readily gave ultrathin sections, which required staining by uranyl acetate and/or lead stains to give adequate contrast for electron microscopy. The ultrastructure of pancreas, kidney, muscle, and intestine was well preserved by aldehyde fixation alone. Use of postfixation in osmium tetroxide or direct osmium tetroxide fixation was unsatisfactory. The fine structure of aldehyde-fixed liver from fasted rats was well preserved, whereas that from normal rats showed considerable disorganization and collapse, apparently because of extraction of glycogen during the embedding procedure. Enzymatic extraction of proteins by pepsin and of ribonucleic acid by ribonuclease after either formaldehyde or glutaraldehyde fixation was rapidly effected by direct treatment of ultrathin sections with solutions of the enzymes. In contrast, no digestion of chromatin by deoxyribonuclease could be detected. In spite of this present limitation, HPMA appears to have several advantages over earlier water-miscible embedding media for electron microscopy and to be particularly suitable for ultrastructural cytochemistry.     

High-resolution light microscopy (HRLM) and digital analysis of Pompe disease pathology.

Pompe disease is an autosomal recessive lysosomal storage disorder caused by a deficiency of the lysosomal enzyme acid alpha-glucosidase, responsible for the degradation of lysosomal glycogen. Absent or low levels of the enzyme leads to lysosomal glycogen accumulation in cardiac and skeletal muscle cells, resulting in progressive muscle weakness and death from cardiac or respiratory failure. Recombinant enzyme replacement and gene therapy are now being investigated as treatment modalities for this disease. A knockout mouse model for Pompe disease, induced by the disruption of exon 6 within the acid alpha-glucosidase gene, mimics the human disease and has been used to evaluate the efficacy of treatment modalities for clearing glycogen. However, for accurate histopathological assessment of glycogen clearance, maximal preservation of in situ lysosomal glycogen is essential. To improve retention of glycogen in Pompe tissues, several fixation and embedding regimens were evaluated. The best glycogen preservation was obtained when tissues fixed with 3% glutaraldehyde and postfixed with 1% osmium tetroxide were processed into epon-araldite. Preservation was confirmed by staining with the Periodic acid-Schiff's reaction and by electron microscopy. This methodology resulted in high-resolution light microscopy (HRLM) sections suitable for digital quantification of glycogen content in heart and skeletal muscle. Combining this method of tissue fixation with computer-assisted histomorphometry has provided us with what we believe is the most objective and reproducible means of evaluating histological glycogen load in Pompe disease.