On the chemistry of formaldehyde fixation and its effects on immunohistochemical reactions.

Formalin has been recommended as an innocuous fixative for immunohistochemistry. However, several studies demonstrated impairment or blocking of antigenic activity of certain proteins. Formalin fixation was discovered accidentally by F. Blum in 1893 and its deleterious effects on various tissue structures were discussed extensively during the following decades. More recently, some authors assumed that formaldehyde bound to tissues can be largely or completely removed by washing and dehydration. According to chemical data, formaldehyde forms highly reactive methylols with uncharged amino groups. Such methylol groups yield methylene bridges with suitably spaced amides, arginine and aromatic amino acid sidechains. Only loosely bound formaldehyde is removed by washing for several hours. Residual bound formaldehyde cannot be dislodged by washing for weeks, but some formaldehyde is gradually removed when tissues are stored in water for an extended number of years. Methylene crosslinks resist treatment with high concentrations of urea, and can be broken only by drastic hydrolysis. It appears unlikely that such firmly bound formaldehyde is removed by conventional washing and dehydration procedures used in histochemistry. The superiority of methacarn, alcohol or acetone over formaldehyde fixation for immunohistochemical demonstration of prekeratin, myosin, type I and type IV collagen, laminin and fibronectin can be ascribed to the irreversible alterations of tissue proteins by formaldehyde.

Orcein, collastin and pseudo-elastica: a re-investigation of Unna's concepts.

Orcein has been recommended for identification of elastin. Since other traditional elastica stains proved to be unspecific, it was deemed of interest to determine the selectivity of orcein and to review pertinent literature. Orcein was employed as a textile dye in ancient Egypt and was used for dyeing of wool and silk until the early 20th century. It was introduced into histological technic in 1878 as a stain for cytoplasm. Unna recommended it for demonstration of elastic tissue in 1890 and retracted claims for its specifity in 1894 because orcein colored also certain collagen fibers. Unna suggested the term collastin for collagen fibers which share the affinity of elastin for acid orcein. Other orcein solutions were used as selective stains for collagen. In histochemical studies, the staining properties of resorcin-fuchsin and orcein were very similar; elastin and various collagen fibers were strongly colored. Unna's collastin is apparently identical with the pseudo-elastica described in sections stained with resorcin-fuchsin. Both dyes react with meshworks of fine fibers, embryonic, experimentally or pathologically altered collagens. It is suggested to use the term collastin, instead of pseudo-elastica, for collagenous fibers which bind the traditional elastica stains.

Demonstration of phosphates in calcium deposits: a modification of von Kossa's reaction.

It has been suggested that in von Kóss'as technic silver cations replace calcium bound to phosphate or carbonate groups and are then reduced to black metallic silver during exposure to light. However, in test tube experiments silver phosphate retains its yellow color for days. These differences between reactions of pure calcium phosphates and calcium deposits in tissues were emphasized already by von Kóssa; he regarded only the initial yellow coloration of calcium diagnostic for calcium phosphates and deplored the subsequent blackening caused by organic compounds. Von Kóssa's experiments were easily reproducible. A review of the literature showed that reduction of silver nitrate by organic compounds was well known in the 19th century. For histochemical studies of phosphates it was deemed desirable to avoid the formation of black by-products. Sections of paraffin-embedded human tissues were exposed to solutions of silver nitrate in subdued light or darkness then treated with sodium thiosulfate. Silver phosphate was yellow to yellowish brown; other tissue structures remained colorless. No darkening was observed in sections stored for eight years. Other compounds which form yellow silver salts, e.g. iodides and periodates, are unlikely to occur in paraffin sections of human tissues.

Demonstration of amyloid with Mesitol WLS-Congo Red: application of a textile auxiliary to histochemistry.

Previous histochemical investigations demonstrated similarities in the binding of Congo Red and other direct cotton dyes by amyloid and cellulose. It seemed therefore of interest to determine whether or not the cellulose-like reactivity of amyloid extends also to dye solutions containing an anionic reserving agent. These reagents are used in the dyeing of wool-cellulose (Halbwolle) fabrics to prevent binding of direct cotton dyes by proteins. Mesitol WLS-Congo Red solutions stained amyloid selectively; other tissue structures, except some hyaline deposits in arterioles, remained unstained. The cause of this non-specific reaction could not be determined with certainty. Therefore, the alkaline Congo Red method is recommended for histochemical identification of amyloid. However, the Mesitol WLS-Congo Red technic was very useful for demonstration of amyloid after prolonged storage of tissues in formalin; amyloid in such material showed little or no reactivity with the alkaline Congo Red or the Sirius dye methods. This pilot study indicates that anionic reserving agents can be effectively employed under conditions of histochemical technics.

On Schiff's bases and aldehyde-fuchsin: a review from H. Schiff to R. D. Lillie.

In histochemistry aldehyde-fuchsin is widely regarded as an azomethine compound, though this hypothesis cannot explain the variety of reaction products. Infrared spectroscopy did not show a C = N bond. It was therefore deemed of interest to review chemical studies of aldehyde-fuchsin and other Schiff's bases by Schiff and his contemporaries. Schiff regarded reaction products of low molecular aliphatic aldehydes, e.g. acetaldehyde, and aromatic amines as diarylamines; aldehyde-fuchsin was assigned a 2:3 (dye:aldehyde) formula. These reactions were facilitated by alcohol and HCl. Others suggested condensation of two aldehyde molecules which carried a secondary and a tertiary amine respectively. Eibner proved that these compounds were ethylenes, not azomethines, and contained two secondary amines. Condensation of such bases produced ethylenic polymers, the Schultz's bases. Aromatic aldehydes readily yielded azomethines; aliphatic aldehydes formed --C = N-- bonds only during prolonged heating. These findings are in agreement with recent chemical data. Clearly, the term Schiff's bases is not synonymous with azomethines, but denotes any reaction product of aldehydes and amines. In 1962, Lillie's histochemical studies confirmed "the secondary amine nature of aldehyde aryl amine condensation". Thus, chemical and histochemical studies from Schiff in the 1860's to Lillie in the 1960's indicate that aldehyde-fuchsin is not an azomethine compound, but contains diarylamines and their derivatives.

A methanol resorcin-fuchsin stain for elastic tissues and nuclei.

The staining properties of conventional ethanol resorcin-fuchsin and of methanol resorcin-fuchsin were compared. Formula; Dissolve 0.2 g of commercial resorcin-fuschin in 70 ml of methanol or ethanol, add 30 ml of water and 1 m1 of concentrated HC1; stain sections for 4 hours. Both solutions colored elastic and pseudoelastic fibers, cartilage and some mucins. Methanol resorcin-fuchsin also colored nuclei in methacarn- (methanol-chloroform-glacial acetic acid 6:3:1) and formalin-fixed tissues; this nuclear stain withstood counterstaining with picro-dye mictures. Zenker-fixed sections showed diffuse coloration with little or no contrast between nuclei and cytoplasm. Extraction with hot trichloracetic acid abolished binding of methylene blue, but binding of methanol resorcin-fuchsin by nuclei remained unaltered or was enhanced. Experiments with solvents containing various concentrations of methanol, ethanol or isopropanol indicated that the staining patterns of resorcin-fuchsin are determined by the nature and concentration of the alcohol. Methanol resorcin-fuchsin proved useful for simultaneous visualization of elastic tissues and nuclei.

Mallory bodies: lesions of hepatocytes containing proteins of the keratin-myosin-epidermin group.

Mallory's alcoholic hyalin in hepatocytes was found also in other diseases and is now referred to as Mallory bodies. Data concerning their histochemical, immuno and electron microscopic properties are partly contradictory. In this study, early stages of Mallory bodies reacted strongly with configurational technics for myosins; affinity tended to decrease when material with the properties of keratohyalin and the matrix of stratum corneum was formed. Thus, many Mallory bodies contained histochemically distinct myoid and keratin-like proteins. Electron microscopists demonstrated thick and thin filaments resembling contractile systems in Mallory bodies; the failure of immunologists to visualize actomyosin may be due to the heterogeneity of these proteins. The currently popular term prekeratin has been applied to a variety of substances extracted from epidermis, hoof and hair under different conditions. The prekeratin of recent immunofluorescence studies seems to contain mainly epidermin and low molecular matrix proteins; both were studied extensively by chemists. Epithelial filaments, including tonofibrils and contractile fibrils regarded as a subgroup of myofibrils, were well known half a century ago, but were banished by electron microscopy. Observations in this study and data on epidermal actomyosin indicate that different proteins of the k-m-e-f group can indeed coexist in epithelial cells. The formation and resolution of Mallory bodies can be regarded as an example of the well known shifts of epithelial cells between secretory and keratinizing states.

Are picro-dye reactions for collagens quantitative? Chemical and histochemical considerations.

Previous studies of picro-dye reactions demonstrated wide variations in the binding of different dyes. Picro-Sirius Red F3BA was recommended because it colors all collagens intensely and is suitable for polarization microscopy. Recent publications on quantitative uses of this stain were surprising. To obtain further information on the chemical mechanisms of dye binding by proteins, 94 sulfonated azo dyes were tested under the conditions of the picro-Sirius Red F3BA reaction. Reaction patterns varied widely, from failure to compete successfully with picrate ions for binding sites to strong coloration of all tissue structures. Only a few dyes stained collagen, reticulum fibers and basement membranes intensely and selectively. The reactivity of dyes was determined by their molecular configuration and the nature and position of substituents. Correlation with physico-chemical data showed that dye binding is due to non-ionic interactions, i.e. van der Waals and dispersion forces and hydrophobic bonding. Coulomb forces do not impart affinity - increasing sulfonation actually decreases dye uptake - but draw dyes within reach of non-ionic sites. Bound dyes form aggregates with additional dye ions; the aggregation number can range from 2 to many powers of 10. Clearly, dye binding by proteins is not stoichiometric.

Current chemical concepts of acids and bases and their application to anionic ("acid") and cationic ("basic") dyes.

In biomedical studies, dyes are divided into "acid" and "basic" dyes. This classification cannot be reconciled with current chemical definitions of acids and bases. Brönsted-Lowry acids are compounds that can donate protons; bases are proton acceptors. The definition of acids and bases is independent of the electric charge, i.e. acids and bases can be neutral, anionic or cationic. Reactions between acids and bases result in formation of new acid-base pairs. Lewis acids and bases do not depend on a particular element, but are characterized by their electronic configurations. Lewis bases are electron donors; Lewis acids are electron acceptors. This classification is also unrelated to the electric charge. Lewis acids and bases interact by formation of coordinate covalent bonds. In histochemistry and histology, dyes containing -SO3-, -COO- and/or -O- groups are classified as "acid" dyes. However, such compounds are electron pair donors and hence Brönsted-Lowry and Lewis anionic bases. Dyes carrying a positive charge are termed "basic" dyes. Chemically, many cationic dyes are Lewis acids because they can add a base, e.g. OH-, acetate, halides. The hypothesis that transformation of -NH2 into ammonium groups imparts "basic" properties to dyes is untenable; ammonium groups are proton donors and hence acids. Furthermore, conversion of an amino into an ammonium group blocks a lone electron pair and the color of the dye changes drastically, e.g. from violet to green and yellow. It appears therefore highly unlikely that ammonium groups are responsible for binding of cationic ("basic") dyes. In histochemistry, it is usually not of critical importance whether anionic or cationic dyes are chemically acids or bases.(ABSTRACT TRUNCATED AT 250 WORDS)

A review of light, polarization and fluorescence microscopic methods for amyloid.

Traditional technics for amyloid are not always dependable. Therefore, several reactions, which do not require differentiation, were developed in this laboratory. This review describes technics that proved suitable for diagnostic pathology, including polarization and fluorescence microscopy. Effects of fixation on the reactivity of amyloid are also considered. The chemical mechanism of the alkaline Congo red, Mesitol WLS-Congo red and Phorwhite BBU reaction, and of a modified thioflavine T stain are reviewed briefly. Problems encountered with other methods, which cannot be recommended for diagnostic pathology, are outlined.

Aldehyde-fuchsin: historical and chemical considerations.

The staining mechanisms of Gomori's aldehyde-fuchsin are not yet fully understood. It seemed therefore timely to review the history of this dye class in context with current dye and aldehyde chemistry. In 1861 Lauth treated basic fuchsin with acetaldehyde. This dye became known as Aldehyde Blue, but consisted of violet and blue dyes. Schiff (1866) studied several aldehyde-fuchsins; these compounds contained two molecules of dye and three molecules of aldehyde. Acetaldehyde-fuchsin prepared according to Schiff's directions showed staining properties similar to those of Gomori's aldehyde-fuchsin. This dye class was soon superseded by new dyes more suitable for textile dyeing, and chemical investigations of aldehyde-fuchsins ceased around the turn of the century. Gomori's aldehyde-fuchsin has been regarded as a Schiff base. However, according to chemical data, low molecular aliphatic aldehydes and aromatic amines tend to form condensation products. Correlations of chemical and histochemical observations suggest such processes during aging of dye solutions. Models of dimers and polymers of aldehyde-fuchsin could be built without steric hindrance. The nature of the bonds formed by various components of aldehyde-fuchsin solutions is not clear. However, cystine in proteins, e.g. in basement membranes, apparently does not play a role in the binding of aldehyde-fuchsin by unoxidized Carnoy- or methacarn-fixed sections.

Infrared fluorescence microscopy of stained tissues: principles and technic.

Infrared photomicrography was used extensively from 1927 to the 1940's, but received little attention during the last decades. However, studies of infrared fluorescence of stained sections could not be found in the accessible literature. Ramsley (1968) published quantitative data on infrared fluorescence of approximately 250 dyes bound to textile fibers. The intensity of infrared fluorescence of many dyes varied widely with the substrate. It was therefore deemed of interest to determine whether or not similar differences in infrared fluorescence may occur when dyes are bound to histochemically distinct tissue structures. Myofibrils and collagens stained with triarylmethane dyes were chosen as test objects. Kodak infrared cut-off filter No. 301 and Wratten filter #16 were used as exciter filters to remove infrared and UV-blue and the light of a xenon lamp. Wratten filter #70 and #89B were employed as barrier filters. Infrared radiation was recorded with Kodak Ektachrome infrared film. To facilitate correlation of infrared fluorescence patterns with visible images, tissues were photographed also with conventional color film. Stained myofibrils, e.g. in myoepithelium, smooth and striated muscle emitted strong infrared fluorescence; collagen showed little or no fluorescence. Barrier filter Wratten #70 permitted simultaneous demonstration of infrared fluorescence and of non-fluorescent structures and thus facilitated histopathological studies. Preliminary findings indicate decrease or loss of infrared fluorescence of stained muscle fibers in various lesions, e.g. myocardial infarction, Duchenne-type muscular dystrophy.

Application of thiazole dyes to amyloid under conditions of direct cotton dyeing: correlation of histochemical and chemical data.

The fluorescent brightening agent Phorwhite (Blankophor) BBU imparts intense selective fluorescence to amyloid, but this modern reagent is no longer readily available on the biological dye market. Conventional Thioflavine S and T stains require differentiation and are not specific. To improve selectivity, direct and cationic thiazole dyes were substituted in the alkaline Congo Red and the Phorwhite BBU procedure. With the former technic Diphenyl Brilliant Yellow 8G, Clayton Yellow, Thiazol Yellow, Thioflavine T and Seto Flavine T imparted strong to intense selective fluorescence to amyloid. Under the conditions of the Phorwhite BBU reaction these dyes were suitable only for formalin-fixed amyloid. Several thiazole dyes did not fluoresce. Fluorescence is a function of the molecular orbital system, the thiazole rings per se cannot induce fluorescence. Paper chromatograms indicated two or more fractions in the dyes studied. Different samples of the same dye can vary significantly in their staining and fluorescence properties. This heterogeneity is inherent in the mode of synthesis. In some cases the cationic thiazole dyes rendered certain amyloid deposits, e.g. in vessel walls, intensely fluorescent; other amyloid deposits in the same sections showed only weak fluorescence. Further studies are required to correlate these peculiar patterns with immunological data on amyloid types.

Application of current chemical concepts to metal-hematein and -brazilein stains.

Current chemical concepts were applied to Weigert's, M. Heidenhain's and Verhoeff's iron hemateins, Mayer's acid hemalum stain and the corresponding brazilein compounds. Fe bonds tightly to oxygen in preference to nitrogen and is unlikely to react with lysyl and arginyl groups of proteins. Binding of unoxidized hematoxylin by various substrates has long been known to professional dyers and was ascribed to hydrogen bonding. Chemical data on the uptake of phenols support this theory. Molecular models indicate a nonplanar configuration of hematoxylin and brazilin. The traditional quinonoid formula of hematein and brazilein was revised. During chelate formation each of the two oxy- groups of the dye shares an electron pair with the metal and contributes a negative charge to the chelate. Consequently, the blue or black 2:1 (dye:metal) complexes are anionic. Olation of such chelates affects the staining properties of iron hematein solutions. The color changes upon oxidation of hematoxylin, reaction of hematein with metals, and during exposure of chelates to acids can be explained by molecular orbital theory. Without differentiation or acid in dye chelate solutions, staining patterns are a function of the metal. Reactions of acidified solutions are determined by the affinities of the dye ligands. Brazilein is much more acid-sensitive than hematein. This difference can be ascribed to the lack of a second free phenolic -OH group in brazilein, i.e. one hydrogen bond is insufficient to anchor the dye to tissues. Since hematein and brazilein are identical in all other respects, their differences in affinity cannot be explained by van der Waals, electrostatic, hydrophobic or other forces.

On structural formulas of basic fuchsin and aldehyde-Schiff reaction products.

A variety of structural formulas has been suggested for the basic fuchsin moiety of aldehyde-Schiff reaction products. It was therefore deemed of interest to review the development of these different concepts of dye structure. Formulation of basic fuchsin as an ammonium salt preceded the quinonoid theory; however, chemists could not find such salts. The quinoid theory also could not be reconciled with chemical observations and spectroscopic data. In the 1920's the quinonoid formulas were superseded by benzenoid formulas with a positive charge at the central carbon atom. Using the resonance theory, basic fuchsin is often written with an apparently pentavalent nitrogen atom at a quinonoid ring. But such limiting structures do not exist in reality; the structure of the resonance hybrid is intermediate between the various contributing structures. The carbonium formula appears preferable to other limiting structures because it would remove temptations to endow basic fuchsin with a quinonoid ring, an imino or an ammonium group. Some formulas of aldehyde-Schiff reaction products carry two positive charges. But divalent basic fuchsin is very unstable. The divalent form of its derivatives formaldehyde- Schiff's reagent, aldehyde-fuchsin, and Crystal Violet is deep green; with decreasing H-ion concentration the divalent green compounds revert to the red or violet monovalent substance. It appears therefore highly unlikely that divalent basic fuchsin exists in PAS reaction products in washed and dehydrated sections.

On the mechanism of Mallory's phosphotungstic acid-haematoxylin stain.

The mechanism of Mallory's (1900) phosphotungstic acid-haematoxylin (PTAH) stain is not yet fully understood; staining properties vary with the age of the dye solution. In this study, a 1 year-old solution was employed as described previously. Paper chromatograms revealed blue, red and yellow components of phosphotungstic acid-haematoxylin. Modification of reactive groups as well as consecutive treatment of sections with haematoxylin and phosphotungstic acid indicated hydrogen bond formation by the blue component. Substitution of brazilin for haematoxylin proved that two free phenolic --OH groups per dye residue are essential for binding of the blue phosphotungstic acid-haematoxylin chelate. Uptake of the red fraction was determined by the polyacid moiety of the dye. Van der Waals forces presumably contributed to dye binding, but did not govern the selectivity of the red and blue components of PTAH for various tissue structures.

Light microscopic distinction between elastin, pseudo-elastica (type III collagen?) AND INTERSTITIAL COLLAGEN.

Distinction between elastin and collagen in arteriosclerotic lesions is difficult because the so-called elastica stains are bound also by collagen fibers which resemble collagen of premature infants. Investigations of effects of organic solvents on dye binding led to the development of methods for selective demonstration of pseudo-elastica, and for simultaneous visualization of elastin and pseudo-elastica in contrasting colors. Paraffin sections of human autopsy material were stained with solutions of resorcin-fuchsin, orcein or aldehyde fuchsin in absolute ethanol. In other series, sections pretreated with this resorcin-fuchsin solution were counter-stained with tannic acid-phosphomolybdic acid (TP)-dye technics. Solutions of these "elastica stains" in absolute ethanol colored only pseudo-elastica; elastin, e.g. elastic membranes of aorta, remained unstained. In sections counterstained with TP-dye technics elastin was colored red; pseudo-elastica retained the purplish blue coloration imparted by resorcin-fuchsin. Other collagens were stained yellow. A review of the literature showed that until the 1920's elastin was classified as a gelatinoid of the collagen group. Elastic fibers were identified by mechanical properties, not a particular chemical composition. Hence, the elastic fibers of classical histology cannot be equated with the elastin of modern chemistry. Correlation of histochemical observations with chemical data indicates that the collagenous pseudo-elastica corresponds to [alpha1(III)]3 collagen.

On the history of basic fuchsin and aldehyde-Schiff reactions from 1862 to 1935.

The nature of products formed by aldehydes and Schiff's reagent, whether they are sulfonic or sulfinic acid compounds, has been the subject of much discussion. It seems therefore timely to review early studies of aldehyde-Schiff reactions, including the history of pararosanilin and related dyes. Dyes of the basic fuchsin group have been studied extensively since 1862, and their triphenylmethane structure was established in 1878. The currently used structural formulas were introduced around the turn of the century. Reactions of basic fuchsin with aldehydes, with and without addition of SO2, were investigated by Schiff in the 1860's i.e. before the structure of these dyes was known. In 1900 Prud'homme showed that the reaction products of basic fuschsin, sodium bisulfite and formaldehyde are alkylated and sulfonated derivatives of the parent compound; further chemical studies indicated attachment of the sulfonic acid group to the carbon atom of the aldehyde. Prud'homme's findings were repeatedly confirmed during the following decades. Wieland and Scheuing were apparently unaware of these studies and introduced the sulfinic acid theory in 1921; furthermore, they considered substitution at two amino group of Schiff's reagent essential for formation of the colored compound. However, later chemical and spectroscopic studies showed no evidence of-N-sulfinic acids but supported the sulfonic acid theory of Prud'homme.

A comparative study of PAS-phosphotungstic acid-Diamine Supra Blue FGL and immunological reactions for type I collagen.

Immunohistochemical techniques proved valuable in histological studies of various types of collagens. However drawbacks include non-specific reactions of antibodies, masking of antigens, and the high cost of antibodies. This study was undertaken to ascertain the specificity of the PAS-phosphotungstic acid-Diamine Supra Blue FGL (PAS-PTA-DSB-FGL) reaction for type I collagen, differentiating it from other collagens. Duplicate series of methacarn-fixed sections of various tissues were treated with the PAS-PTA-DSB FGL reaction and the peroxidase-antiperoxidase (PAP) technique for type I collagen and the staining patterns were compared. Fibers binding the blue dye were found only at sites reacting with antibodies against type I collagen. These observations indicate that the PAS-PTA-DSB FGL procedure is suitable for visualization of type I collagen, e.g. in screening of large series of sections and in the practice of surgical and autopsy pathology.